DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.
Provide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.
A useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.
The producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.
Choose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.
The native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.
For a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.
The producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.
The dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.
The maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.
The 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.
The useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.
", "aboutText": "DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.\n\n### Start with the material and the artifact\n\nProvide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.\n\nA useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.\n\n### Choose direct chat or thinking\n\nThe producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.\n\nChoose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.\n\n### Connect tools through an application workflow\n\nThe native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.\n\nFor a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.\n\nThe producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.\n\n### Budget the conversation and the answer\n\nThe dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.\n\nThe maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.\n\n### Keep the release and the architecture distinct\n\nThe 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.\n\nThe useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.", "editorialOverlay": { "copy_fields": { "practical_html": "Technical example in English
Repair a required-field check\n\nA route rejects only if not payload.get("quantity"). The contract requires quantity to be an integer greater than zero, excluding booleans. Explain why inputs {"quantity": -2} and {"quantity": true} expose a problem. Propose a small Python validation patch and cases for missing, zero, negative, boolean and positive values.A diagnosis, a focused validation patch and five input cases with accept or reject expectations.Check that Python booleans are not accepted as integers and positive whole quantities still pass.deepseek-v4-flash-0731DeepSeek V4 Flash 0731 is the official Flash release for text, coding and agent workflows. It offers direct chat and thinking modes, structured tool-call output and a million-token model context; its dated edition includes the DSpark speculative-decoding module.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.
Provide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.
A useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.
The producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.
Choose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.
The native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.
For a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.
The producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.
The dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.
The maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.
The 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.
The useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.
POST /v1/messagestexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek-v4-flash-0731 | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits Minimum items: 1 | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits user, assistant | Default— |
Fieldmessages[].content | Typestring|content_block[] | RequirementRequired | Values & limits | Default— |
Fieldsystem | Typestring|content_block[] | RequirementOptional | Values & limits | Default— |
Fieldmax_tokens | Typeinteger | RequirementRequired | Values & limits Minimum: 1 Limit applies to: exact selected model/card ceiling | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Fieldtemperature | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtop_p | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldstop_sequences | Typestring[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools | Typetool[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools[].name | Typestring | RequirementOptional | Values & limits Applies when: tool supplied | Default— |
Fieldtools[].description | Typestring | RequirementOptional | Values & limits | Default— |
Fieldtools[].input_schema | TypeJSON_schema | RequirementOptional | Values & limits | Default— |
Fieldtool_choice | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldthinking.type | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldthinking.budget_tokens | Typeinteger | RequirementOptional | Values & limits Model-specific | Default— |
Fieldoutput_config.effort | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldmetadata | Typeobject | RequirementOptional | Values & limits | Default— |
Fieldcache_control | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Text · UTF8 text
messages[].content or type=text block
| Parameter | Contract |
|---|---|
Parametermaterial_types | Contracttext |
Parameterrequired | Contractmodel, messages, max_tokens |
Parameteroptional_controls | Contractnull |
Parametercontent_type | Contractapplication/json |
Parametertext_file_material | Contractactual UTF-8 content inserted as text; file IDs/URLs need separate exact support |
Lifecycle: exact retained published method
Method shapes and IDs are exact; supplier capability is separate from admission.
POST /v1/messagesAiloot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Model-level reference schedules do not change method selection, quotes or charges. Active charges use the selected Ailoot method rates.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationOpenAI Chat · Base token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationOpenAI Chat · Request-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationAnthropic Messages · Base token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationAnthropic Messages · Request-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\n: "${AILOOT_TEXT_FILE:?Set your actual UTF-8 source-material file}"\nexport AILOOT_TEXT_FILE\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "deepseek-v4-flash-0731", "max_tokens": 1024, "messages": []}')\nf=Path(os.environ["AILOOT_TEXT_FILE"])\nif not f.is_file():raise SystemExit("Choose your actual UTF-8 source material")\ntext=f.read_text(encoding="utf-8")\nif not text.strip():raise SystemExit("Source material must contain text")\np["messages"]=[{"role":"user","content":"Summarize this material and quote supporting source passages.\\n\\n"+text}]\nPath("request.json").write_text(json.dumps(p,ensure_ascii=False))\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/messages \\\n -H "x-api-key: $AILOOT_API_KEY" -H 'anthropic-version: 2023-06-01' -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport json\nfrom pathlib import Path\nprint(json.dumps(json.loads(Path("result.json").read_text()),ensure_ascii=False,indent=2))\nPY_RESULT\nimport anthropic\nimport os\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient = anthropic.Anthropic(base_url="https://api.ailoot.ai", api_key=os.environ["AILOOT_API_KEY"], max_retries=0)\nmsg = client.messages.create(\n model="deepseek-v4-flash-0731",\n max_tokens=1024,\n messages=[{"role": "user", "content": "Explain the difference between an observation and a hypothesis in three sentences."}],\n)\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nsave_result(msg.to_dict(mode="json"))\nimport Anthropic from "@anthropic-ai/sdk";\n\nif (!process.env.AILOOT_API_KEY?.trim()) throw new Error("Set AILOOT_API_KEY before making a request");\nconst client = new Anthropic({ baseURL: "https://api.ailoot.ai", apiKey: process.env.AILOOT_API_KEY, maxRetries: 0 });\nconst msg = await client.messages.create({\n model: "deepseek-v4-flash-0731",\n max_tokens: 1024,\n messages: [{ role: "user", content: "Explain the difference between an observation and a hypothesis in three sentences." }],\n});\nimport { writeFileSync } from "node:fs";\n\nfunction saveResult(data, stem = "result") {\n writeFileSync(stem + ".json", JSON.stringify(data, null, 2));\n console.log(JSON.stringify(data, null, 2));\n const saveBlob = (encoded, suffix, label) => {\n const raw = Buffer.from(encoded, "base64");\n const ext = raw.subarray(0,8).equals(Buffer.from([137,80,78,71,13,10,26,10])) ? "png" : raw[0]===255 && raw[1]===216 ? "jpg" : raw.subarray(0,4).toString()==="RIFF" && raw.subarray(8,12).toString()==="WEBP" ? "webp" : suffix;\n const file = `${stem}-${label}.${ext}`;\n writeFileSync(file, raw); console.log("Saved", file);\n };\n for (const [i, block] of (data.content || []).entries()) console.log("Content", i, JSON.stringify(block));\n for (const choice of data.choices || []) console.log("Choice", choice.index, JSON.stringify(choice.message || choice.delta || {}));\n for (const [i, item] of (data.output || []).entries()) {\n console.log("Output", i, JSON.stringify(item));\n if (item.type === "image_generation_call" && item.result) saveBlob(item.result, item.output_format || "bin", `output-${i}`);\n }\n for (const [ci, candidate] of (data.candidates || []).entries()) {\n for (const [pi, part] of (candidate.content?.parts || []).entries()) {\n console.log("Candidate/part", ci, pi, JSON.stringify(part));\n const blob = part.inlineData || part.inline_data;\n if (blob?.data) {\n const mime = blob.mimeType || blob.mime_type;\n const suffix = ({"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"})[mime] || "bin";\n saveBlob(blob.data, suffix, `candidate-${ci}-part-${pi}`);\n }\n }\n }\n}\n\nsaveResult(msg);\nimport anthropic,os\nfrom pathlib import Path\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=anthropic.Anthropic(base_url="https://api.ailoot.ai",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nwith client.messages.stream(model='deepseek-v4-flash-0731',max_tokens=1024,messages=[{"role":"user","content":"Explain observations and hypotheses in three sentences."}]) as stream:\n with Path("stream-events.jsonl").open("w",encoding="utf-8") as log:\n for event in stream:\n payload=event.to_dict(mode="json")\n log.write(json.dumps(payload,ensure_ascii=False)+"\\n"); log.flush()\n print(json.dumps(payload,ensure_ascii=False))\n final=stream.get_final_message()\n save_result(final.to_dict(mode="json"),"stream-result")\njson or SSE
Keep every complete returned response, file and content part.
Keep all content blocks, thinking signatures and actual tool-use IDs. Assemble all SSE block deltas; a first text block is not the complete output.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldid | Typestring |
Fieldmodel | Typestring |
Fieldcontent[] | Typetext|thinking|tool_use|other content block |
Fieldstop_reason | Typestring|null |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot deepseek-v4-flash-0731
Ailoot Ultra deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest
Use the method paths shown above.
deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest, deepseek-v4-flash-0731
deepseek-v4-flash-0731DeepSeek V4 Flash 0731 is the official Flash release for text, coding and agent workflows. It offers direct chat and thinking modes, structured tool-call output and a million-token model context; its dated edition includes the DSpark speculative-decoding module.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.
Provide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.
A useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.
The producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.
Choose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.
The native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.
For a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.
The producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.
The dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.
The maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.
The 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.
The useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.
POST /v1/responsestexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek-v4-flash-0731 | Default— |
Fieldinput | Typestring|input_item[] | RequirementRequired | Values & limits | Default— |
Fieldinstructions | Typestring | RequirementOptional | Values & limits | Default— |
Fieldinput[].type | Typestring | RequirementOptional | Values & limits message, reasoning, function_call, function_call_output | Default— |
Fieldinput[].content | Typestring|content_part[] | RequirementOptional | Values & limits | Default— |
Fieldinput[].call_id | Typestring | RequirementOptional | Values & limits Applies when: function_call_output; retain actual returned call_id | Default— |
Fieldinput[].output | Typestring|content_part[] | RequirementOptional | Values & limits Applies when: function_call_output | Default— |
Fieldmax_output_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Fieldtemperature | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtop_p | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools | Typetool[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtool_choice | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldparallel_tool_calls | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning.effort | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning.summary | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtext.format | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtext.verbosity | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldinclude | Typestring[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldstore | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldprevious_response_id | Typestring | RequirementOptional | Values & limits ID requirement: Only an actual response ID from a supported continuation Model-specific | Default— |
Fieldservice_tier | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldprompt_cache_key | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldprompt_cache_options | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Text · UTF8 text
input text or input_text part
| Parameter | Contract |
|---|---|
Parametermaterial_types | Contracttext |
Parameterrequired | Contractmodel, input |
Parameteroptional_controls | Contractnull |
Parametercontent_type | Contractapplication/json |
Parametertext_file_material | Contractactual UTF-8 content inserted as text; file IDs/URLs need separate exact support |
Lifecycle: exact retained published method
Method shapes and IDs are exact; supplier capability is separate from admission.
POST /v1/responsesAiloot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Model-level reference schedules do not change method selection, quotes or charges. Active charges use the selected Ailoot method rates.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationOpenAI Chat · Base token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationOpenAI Chat · Request-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationAnthropic Messages · Base token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationAnthropic Messages · Request-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\n: "${AILOOT_TEXT_FILE:?Set your actual UTF-8 source-material file}"\nexport AILOOT_TEXT_FILE\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "deepseek-v4-flash-0731", "input": ""}')\nf=Path(os.environ["AILOOT_TEXT_FILE"])\nif not f.is_file():raise SystemExit("Choose your actual UTF-8 source material")\ntext=f.read_text(encoding="utf-8")\nif not text.strip():raise SystemExit("Source material must contain text")\np["input"]="Summarize this material and quote supporting source passages.\\n\\n"+text\nPath("request.json").write_text(json.dumps(p,ensure_ascii=False))\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/responses \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nsave_result(json.loads(Path("result.json").read_text(encoding="utf-8")))\nPY_RESULT\nfrom openai import OpenAI\nimport os\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient = OpenAI(base_url="https://api.ailoot.ai/v1", api_key=os.environ["AILOOT_API_KEY"], max_retries=0)\nresp = client.responses.create(model="deepseek-v4-flash-0731", input="Explain the difference between an observation and a hypothesis in three sentences.")\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nsave_result(resp.to_dict(mode="json"))\nimport OpenAI from "openai";\n\nif (!process.env.AILOOT_API_KEY?.trim()) throw new Error("Set AILOOT_API_KEY before making a request");\nconst client = new OpenAI({ baseURL: "https://api.ailoot.ai/v1", apiKey: process.env.AILOOT_API_KEY, maxRetries: 0 });\nconst resp = await client.responses.create({ model: "deepseek-v4-flash-0731", input: "Explain the difference between an observation and a hypothesis in three sentences." });\nimport { writeFileSync } from "node:fs";\n\nfunction saveResult(data, stem = "result") {\n writeFileSync(stem + ".json", JSON.stringify(data, null, 2));\n console.log(JSON.stringify(data, null, 2));\n const saveBlob = (encoded, suffix, label) => {\n const raw = Buffer.from(encoded, "base64");\n const ext = raw.subarray(0,8).equals(Buffer.from([137,80,78,71,13,10,26,10])) ? "png" : raw[0]===255 && raw[1]===216 ? "jpg" : raw.subarray(0,4).toString()==="RIFF" && raw.subarray(8,12).toString()==="WEBP" ? "webp" : suffix;\n const file = `${stem}-${label}.${ext}`;\n writeFileSync(file, raw); console.log("Saved", file);\n };\n for (const [i, block] of (data.content || []).entries()) console.log("Content", i, JSON.stringify(block));\n for (const choice of data.choices || []) console.log("Choice", choice.index, JSON.stringify(choice.message || choice.delta || {}));\n for (const [i, item] of (data.output || []).entries()) {\n console.log("Output", i, JSON.stringify(item));\n if (item.type === "image_generation_call" && item.result) saveBlob(item.result, item.output_format || "bin", `output-${i}`);\n }\n for (const [ci, candidate] of (data.candidates || []).entries()) {\n for (const [pi, part] of (candidate.content?.parts || []).entries()) {\n console.log("Candidate/part", ci, pi, JSON.stringify(part));\n const blob = part.inlineData || part.inline_data;\n if (blob?.data) {\n const mime = blob.mimeType || blob.mime_type;\n const suffix = ({"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"})[mime] || "bin";\n saveBlob(blob.data, suffix, `candidate-${ci}-part-${pi}`);\n }\n }\n }\n}\n\nsaveResult(resp);\nfrom openai import OpenAI\nimport os\nfrom pathlib import Path\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nwith client.responses.stream(model="deepseek-v4-flash-0731", input="Explain the difference between an observation and a hypothesis in three sentences.") as stream:\n with Path("stream-events.jsonl").open("w",encoding="utf-8") as log:\n for event in stream:\n payload=event.to_dict(mode="json")\n log.write(json.dumps(payload,ensure_ascii=False)+"\\n"); log.flush()\n print(json.dumps(payload,ensure_ascii=False))\n final=stream.get_final_response()\n save_result(final.to_dict(mode="json"),"stream-result")\njson or SSE
Keep every complete returned response, file and content part.
Keep every output item and content part, actual function call_id, encrypted reasoning and usage. In streaming retain complete event/item identity.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldid | Typestring |
Fieldstatus | Typestring |
Fieldoutput[] | Typemessage|reasoning|function_call|other item |
Fieldoutput[].content[] | Typecontent part |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot deepseek-v4-flash-0731
Ailoot Ultra deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest
Use the method paths shown above.
deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest, deepseek-v4-flash-0731
deepseek-v4-flash-0731DeepSeek V4 Flash 0731 is the official Flash release for text, coding and agent workflows. It offers direct chat and thinking modes, structured tool-call output and a million-token model context; its dated edition includes the DSpark speculative-decoding module.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.
Provide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.
A useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.
The producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.
Choose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.
The native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.
For a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.
The producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.
The dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.
The maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.
The 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.
The useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek-v4-flash-0731 | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits Minimum items: 1 | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits Examples: system, user, assistant, tool, function Conditions: These roles are explicitly represented by Source compatibility types; the exact model role set remains model-specific. Model-specific | Default— |
Fieldmessages[].content | Typestring|content_part[]|null | RequirementOptional | Values & limits | Default— |
Fieldmessages[].tool_calls | Typetool_call[] | RequirementOptional | Values & limits | Default— |
Fieldmessages[].tool_call_id | Typestring | RequirementOptional | Values & limits Applies when: role=tool; use actual returned ID | Default— |
Fieldmessages[].name | Typestring | RequirementOptional | Values & limits | Default— |
Fieldmessages[].reasoning_content | Typestring | RequirementOptional | Values & limits Applies when: Round-trip actual returned reasoning where this model uses it | Default— |
Fieldmessages[].reasoning_details | Typeobject|array | RequirementOptional | Values & limits Model-specific | Default— |
Fieldmax_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldmax_completion_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Fieldstream_options.include_usage | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtemperature | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtop_p | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldstop | Typestring|string[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools | Typefunction_tool[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools[].type | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools[].function.name | Typestring | RequirementRequired | Values & limits Applies when: function tool supplied | Default— |
Fieldtools[].function.description | Typestring | RequirementOptional | Values & limits | Default— |
Fieldtools[].function.parameters | TypeJSON_schema | RequirementOptional | Values & limits | Default— |
Fieldtool_choice | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldparallel_tool_calls | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldresponse_format | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning_effort | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldservice_tier | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldinstructions | Typestring | RequirementOptional | Values & limits Compatibility field | Default— |
Fieldfunctions | Typefunction[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldfunction_call | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldprompt_cache_options | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Text · UTF8 text
messages[].content
| Parameter | Contract |
|---|---|
Parametermaterial_types | Contracttext |
Parameterrequired | Contractmodel, messages |
Parameteroptional_controls | Contractnull |
Parametercontent_type | Contractapplication/json |
Parametertext_file_material | Contractactual UTF-8 content inserted as text; file IDs/URLs need separate exact support |
Lifecycle: exact retained published method
Method shapes and IDs are exact; supplier capability is separate from admission.
POST /v1/chat/completionsAiloot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
Model-level reference schedules do not change method selection, quotes or charges. Active charges use the selected Ailoot method rates.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationOpenAI Chat · Base token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationOpenAI Chat · Request-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationAnthropic Messages · Base token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationAnthropic Messages · Request-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\n: "${AILOOT_TEXT_FILE:?Set your actual UTF-8 source-material file}"\nexport AILOOT_TEXT_FILE\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "deepseek-v4-flash-0731", "messages": []}')\nf=Path(os.environ["AILOOT_TEXT_FILE"])\nif not f.is_file():raise SystemExit("Choose your actual UTF-8 source material")\ntext=f.read_text(encoding="utf-8")\nif not text.strip():raise SystemExit("Source material must contain text")\np["messages"]=[{"role":"user","content":"Summarize this material and quote supporting source passages.\\n\\n"+text}]\nPath("request.json").write_text(json.dumps(p,ensure_ascii=False))\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/chat/completions \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport json\nfrom pathlib import Path\nprint(json.dumps(json.loads(Path("result.json").read_text()),ensure_ascii=False,indent=2))\nPY_RESULT\nfrom openai import OpenAI\nimport os\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient = OpenAI(base_url="https://api.ailoot.ai/v1", api_key=os.environ["AILOOT_API_KEY"], max_retries=0)\nresp = client.chat.completions.create(\n model="deepseek-v4-flash-0731",\n messages=[{"role": "user", "content": "Explain the difference between an observation and a hypothesis in three sentences."}],\n)\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nsave_result(resp.to_dict(mode="json"))\nimport OpenAI from "openai";\n\nif (!process.env.AILOOT_API_KEY?.trim()) throw new Error("Set AILOOT_API_KEY before making a request");\nconst client = new OpenAI({ baseURL: "https://api.ailoot.ai/v1", apiKey: process.env.AILOOT_API_KEY, maxRetries: 0 });\nconst resp = await client.chat.completions.create({\n model: "deepseek-v4-flash-0731",\n messages: [{ role: "user", content: "Explain the difference between an observation and a hypothesis in three sentences." }],\n});\nimport { writeFileSync } from "node:fs";\n\nfunction saveResult(data, stem = "result") {\n writeFileSync(stem + ".json", JSON.stringify(data, null, 2));\n console.log(JSON.stringify(data, null, 2));\n const saveBlob = (encoded, suffix, label) => {\n const raw = Buffer.from(encoded, "base64");\n const ext = raw.subarray(0,8).equals(Buffer.from([137,80,78,71,13,10,26,10])) ? "png" : raw[0]===255 && raw[1]===216 ? "jpg" : raw.subarray(0,4).toString()==="RIFF" && raw.subarray(8,12).toString()==="WEBP" ? "webp" : suffix;\n const file = `${stem}-${label}.${ext}`;\n writeFileSync(file, raw); console.log("Saved", file);\n };\n for (const [i, block] of (data.content || []).entries()) console.log("Content", i, JSON.stringify(block));\n for (const choice of data.choices || []) console.log("Choice", choice.index, JSON.stringify(choice.message || choice.delta || {}));\n for (const [i, item] of (data.output || []).entries()) {\n console.log("Output", i, JSON.stringify(item));\n if (item.type === "image_generation_call" && item.result) saveBlob(item.result, item.output_format || "bin", `output-${i}`);\n }\n for (const [ci, candidate] of (data.candidates || []).entries()) {\n for (const [pi, part] of (candidate.content?.parts || []).entries()) {\n console.log("Candidate/part", ci, pi, JSON.stringify(part));\n const blob = part.inlineData || part.inline_data;\n if (blob?.data) {\n const mime = blob.mimeType || blob.mime_type;\n const suffix = ({"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"})[mime] || "bin";\n saveBlob(blob.data, suffix, `candidate-${ci}-part-${pi}`);\n }\n }\n }\n}\n\nsaveResult(resp);\nfrom openai import OpenAI\nimport os\nfrom pathlib import Path\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nwith client.chat.completions.stream(model='deepseek-v4-flash-0731',messages=[{"role":"user","content":"Explain observations and hypotheses in three sentences."}]) as stream:\n with Path("stream-events.jsonl").open("w",encoding="utf-8") as log:\n for event in stream:\n payload=event.to_dict(mode="json")\n log.write(json.dumps(payload,ensure_ascii=False)+"\\n"); log.flush()\n print(json.dumps(payload,ensure_ascii=False))\n final=stream.get_final_completion()\n save_result(final.to_dict(mode="json"),"stream-result")\njson or SSE
Keep every complete returned response, file and content part.
Keep every choice and its content/tool calls/reasoning. In SSE assemble deltas by choice/tool index, retain the final finish reason and usage.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldchoices[].index | Typeinteger |
Fieldchoices[].message.content | Typestring|null |
Fieldchoices[].message.tool_calls | Typetool_call[] |
Fieldchoices[].message.reasoning_content | Typestring |
Fieldchoices[].message.reasoning_details | Typeobject|array |
Fieldchoices[].finish_reason | Typestring|null |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot deepseek-v4-flash-0731
Ailoot Ultra deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest
Use the method paths shown above.
deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest, deepseek-v4-flash-0731
deepseek/deepseek-v4-flash-0731DeepSeek V4 Flash 0731 is the official Flash release for text, coding and agent workflows. It offers direct chat and thinking modes, structured tool-call output and a million-token model context; its dated edition includes the DSpark speculative-decoding module.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.
Provide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.
A useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.
The producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.
Choose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.
The native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.
For a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.
The producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.
The dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.
The maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.
The 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.
The useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek/deepseek-v4-flash-0731 | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits Minimum items: 1 | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits Examples: system, user, assistant, tool, function Conditions: These roles are explicitly represented by Source compatibility types; the exact model role set remains model-specific. Model-specific | Default— |
Fieldmessages[].content | Typestring|content_part[]|null | RequirementOptional | Values & limits | Default— |
Fieldmessages[].tool_calls | Typetool_call[] | RequirementOptional | Values & limits | Default— |
Fieldmessages[].tool_call_id | Typestring | RequirementOptional | Values & limits Applies when: role=tool; use actual returned ID | Default— |
Fieldmessages[].name | Typestring | RequirementOptional | Values & limits | Default— |
Fieldmessages[].reasoning_content | Typestring | RequirementOptional | Values & limits Applies when: Round-trip actual returned reasoning where this model uses it | Default— |
Fieldmessages[].reasoning_details | Typeobject|array | RequirementOptional | Values & limits Model-specific | Default— |
Fieldmax_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldmax_completion_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Fieldstream_options.include_usage | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtemperature | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtop_p | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldstop | Typestring|string[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools | Typefunction_tool[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools[].type | Typestring | RequirementOptional | Values & limits function Model-specific | Default— |
Fieldtools[].function.name | Typestring | RequirementRequired | Values & limits Applies when: function tool supplied | Default— |
Fieldtools[].function.description | Typestring | RequirementOptional | Values & limits | Default— |
Fieldtools[].function.parameters | TypeJSON_schema | RequirementOptional | Values & limits | Default— |
Fieldtool_choice | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldparallel_tool_calls | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldresponse_format | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning_effort | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldservice_tier | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldinstructions | Typestring | RequirementOptional | Values & limits Compatibility field | Default— |
Fieldfunctions | Typefunction[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldfunction_call | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldprompt_cache_options | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldmodalities | Typestring[] | RequirementOptional | Values & limits text | Default— |
Fieldprovider | Typeobject | RequirementOptional | Values & limits Conditions: provider.options is rejected on this Ultra text route | Default— |
Fieldcache_control.ttl | Typestring | RequirementOptional | Values & limits Conditions: 1h is rejected in actual prompt-cache locations | Default— |
Text · UTF8 text
messages[].content
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages |
Parametermaterial_types | Contracttext |
Parametermessages | Contractmessage array with role/content shape |
Parameterfile_material | Contractactual UTF-8 document inserted as literal text; no file ID inferred |
Parameteroptional_controls | Contractstream, temperature, top_p, stop, max_tokens, max_completion_tokens, reasoning, response_format, tools, tool_choice |
Parametercontrol_scope | Contractmodel-specific; user function tools differ from hosted tools |
Parameteroutput_modalities | Contracttext |
Parameterhosted_operations | Contractfalse |
Parametermultimodal_input_binding | Contractnull |
Lifecycle: synchronous_or_streamed
Send exact model ID with messages.
Context length is not an output cap.
A text file may supply literal text; image/audio/file-object inputs need exact binding.
Tariff references do not add request options. Supported parameters are listed for the selected method.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationRequest-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "deepseek/deepseek-v4-flash-0731", "messages": [{"role": "user", "content": "Explain the difference between an observation and a hypothesis in three sentences."}]}')\n\nPath("request.json").write_text(json.dumps(p,ensure_ascii=False))\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/chat/completions \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport json\nfrom pathlib import Path\nr=json.loads(Path("result.json").read_text())\nfor x in r.get("choices",[]):\n if x.get("message",{}).get("content"): print(x["message"]["content"])\nPY_RESULT\nfrom openai import OpenAI\nimport os\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nresult=client.chat.completions.create(model='deepseek/deepseek-v4-flash-0731',messages=[{"role":"user","content":"Explain observations and hypotheses in three sentences."}])\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nsave_result(result.to_dict(mode="json"))\nfrom openai import OpenAI\nimport os\nfrom pathlib import Path\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nwith client.chat.completions.stream(model='deepseek/deepseek-v4-flash-0731',messages=[{"role":"user","content":"Explain observations and hypotheses in three sentences."}]) as stream:\n with Path("stream-events.jsonl").open("w",encoding="utf-8") as log:\n for event in stream:\n payload=event.to_dict(mode="json")\n log.write(json.dumps(payload,ensure_ascii=False)+"\\n"); log.flush()\n print(json.dumps(payload,ensure_ascii=False))\n final=stream.get_final_completion()\n save_result(final.to_dict(mode="json"),"stream-result")\njson or SSE
Keep every complete returned response, file and content part.
Keep every choice and its content/tool calls/reasoning. In SSE assemble deltas by choice/tool index, retain the final finish reason and usage.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldchoices[].index | Typeinteger |
Fieldchoices[].message.content | Typestring|null |
Fieldchoices[].message.tool_calls | Typetool_call[] |
Fieldchoices[].message.reasoning_content | Typestring |
Fieldchoices[].message.reasoning_details | Typeobject|array |
Fieldchoices[].finish_reason | Typestring|null |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot deepseek-v4-flash-0731
Ailoot Ultra deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest
Use the method paths shown above.
deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest, deepseek-v4-flash-0731
deepseek/deepseek-v4-flash-0731DeepSeek V4 Flash 0731 is the official Flash release for text, coding and agent workflows. It offers direct chat and thinking modes, structured tool-call output and a million-token model context; its dated edition includes the DSpark speculative-decoding module.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.
Provide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.
A useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.
The producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.
Choose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.
The native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.
For a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.
The producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.
The dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.
The maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.
The 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.
The useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek/deepseek-v4-flash-0731 | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits user, assistant | Default— |
Fieldmessages[].content | Typestring | RequirementRequired | Values & limits | Default— |
Fieldmax_tokens | Typeinteger | RequirementRequired | Values & limits | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Text · UTF-8 text
messages[].content string
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages, max_tokens |
Parameteroptional_controls | Contractstream |
Parametermaterial_types | Contracttext |
Parametertext_file_material | ContractRead the actual UTF-8 document and put its text in the indicated input or message content string. |
Lifecycle: synchronous_or_streamed
Tariff references do not add request options. Supported parameters are listed for the selected method.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationRequest-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "deepseek/deepseek-v4-flash-0731",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\n: "${AILOOT_TEXT_FILE:?Set your actual UTF-8 document}"\nexport AILOOT_TEXT_FILE\npython3 - <<'PY_REQUEST'\nimport json, os\nfrom pathlib import Path\n\nsource_file = Path(os.environ['AILOOT_TEXT_FILE'])\nif not source_file.is_file():\n raise SystemExit('Set AILOOT_TEXT_FILE to your actual UTF-8 document')\ntext = source_file.read_text(encoding='utf-8')\nif not text.strip():\n raise SystemExit('Choose a document containing text')\npayload = {'model': 'deepseek/deepseek-v4-flash-0731', 'max_tokens': 128, 'messages': [{'role': 'user', 'content': 'Summarize the incident report.'}]}\npayload['messages'][0]['content'] = 'Summarize this material and quote supporting source passages.\\n\\n' + text\nPath('request.json').write_text(json.dumps(payload, ensure_ascii=False), encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/messages \\\n -H "x-api-key: $AILOOT_API_KEY" -H 'anthropic-version: 2023-06-01' -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport json\nfrom pathlib import Path\n\nresult = json.loads(Path('result.json').read_text(encoding='utf-8'))\nprint(json.dumps(result, ensure_ascii=False, indent=2))\nPY_RESULT\nimport json, os\nfrom pathlib import Path\n\nsource_file = Path(os.environ['AILOOT_TEXT_FILE'])\nif not source_file.is_file():\n raise SystemExit('Set AILOOT_TEXT_FILE to your actual UTF-8 document')\ntext = source_file.read_text(encoding='utf-8')\nif not text.strip():\n raise SystemExit('Choose a document containing text')\npayload = {'model': 'deepseek/deepseek-v4-flash-0731', 'max_tokens': 128, 'messages': [{'role': 'user', 'content': 'Summarize the incident report.'}]}\npayload['messages'][0]['content'] = 'Summarize this material and quote supporting source passages.\\n\\n' + text\nfrom urllib.request import Request, urlopen\nfrom urllib.error import HTTPError\n\napi_key = os.environ['AILOOT_API_KEY']\nif not api_key.strip():\n raise SystemExit('Set AILOOT_API_KEY before making a request')\nheaders = {'x-api-key': 'PLACEHOLDER_KEY', 'anthropic-version': '2023-06-01', 'Content-Type': 'application/json'}\nheaders['x-api-key'] = api_key\nrequest = Request('https://api.ailoot.ai/v1/messages', data=json.dumps(payload, ensure_ascii=False).encode('utf-8'), headers=headers, method='POST')\ntry:\n with urlopen(request, timeout=120) as response:\n raw = response.read()\n Path('response-headers.txt').write_text(str(response.headers), encoding='utf-8')\nexcept HTTPError as error:\n Path('result.json').write_bytes(error.read())\n Path('response-headers.txt').write_text(str(error.headers), encoding='utf-8')\n raise\nPath('result.json').write_bytes(raw)\nimport json\nfrom pathlib import Path\n\nresult = json.loads(Path('result.json').read_text(encoding='utf-8'))\nprint(json.dumps(result, ensure_ascii=False, indent=2))\nJSON or SSE
Keep every complete returned response, file and content part.
Keep all content blocks, thinking signatures and actual tool-use IDs. Assemble all SSE block deltas; a first text block is not the complete output.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldid | Typestring |
Fieldmodel | Typestring |
Fieldcontent[] | Typetext|thinking|tool_use|other content block |
Fieldstop_reason | Typestring|null |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot deepseek-v4-flash-0731
Ailoot Ultra deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest
Use the method paths shown above.
deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest, deepseek-v4-flash-0731
~deepseek/deepseek-v4-flash-latestDeepSeek V4 Flash 0731 is the official Flash release for text, coding and agent workflows. It offers direct chat and thinking modes, structured tool-call output and a million-token model context; its dated edition includes the DSpark speculative-decoding module.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.
Provide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.
A useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.
The producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.
Choose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.
The native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.
For a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.
The producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.
The dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.
The maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.
The 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.
The useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits ~deepseek/deepseek-v4-flash-latest | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits Minimum items: 1 | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits Examples: system, user, assistant, tool, function Conditions: These roles are explicitly represented by Source compatibility types; the exact model role set remains model-specific. Model-specific | Default— |
Fieldmessages[].content | Typestring|content_part[]|null | RequirementOptional | Values & limits | Default— |
Fieldmessages[].tool_calls | Typetool_call[] | RequirementOptional | Values & limits | Default— |
Fieldmessages[].tool_call_id | Typestring | RequirementOptional | Values & limits Applies when: role=tool; use actual returned ID | Default— |
Fieldmessages[].name | Typestring | RequirementOptional | Values & limits | Default— |
Fieldmessages[].reasoning_content | Typestring | RequirementOptional | Values & limits Applies when: Round-trip actual returned reasoning where this model uses it | Default— |
Fieldmessages[].reasoning_details | Typeobject|array | RequirementOptional | Values & limits Model-specific | Default— |
Fieldmax_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldmax_completion_tokens | Typeinteger | RequirementOptional | Values & limits Minimum: 1 Model-specific | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Fieldstream_options.include_usage | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtemperature | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtop_p | Typenumber | RequirementOptional | Values & limits Model-specific | Default— |
Fieldstop | Typestring|string[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools | Typefunction_tool[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldtools[].type | Typestring | RequirementOptional | Values & limits function Model-specific | Default— |
Fieldtools[].function.name | Typestring | RequirementRequired | Values & limits Applies when: function tool supplied | Default— |
Fieldtools[].function.description | Typestring | RequirementOptional | Values & limits | Default— |
Fieldtools[].function.parameters | TypeJSON_schema | RequirementOptional | Values & limits | Default— |
Fieldtool_choice | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldparallel_tool_calls | Typeboolean | RequirementOptional | Values & limits Model-specific | Default— |
Fieldresponse_format | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning_effort | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldreasoning | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldservice_tier | Typestring | RequirementOptional | Values & limits Model-specific | Default— |
Fieldinstructions | Typestring | RequirementOptional | Values & limits Compatibility field | Default— |
Fieldfunctions | Typefunction[] | RequirementOptional | Values & limits Model-specific | Default— |
Fieldfunction_call | Typestring|object | RequirementOptional | Values & limits Model-specific | Default— |
Fieldprompt_cache_options | Typeobject | RequirementOptional | Values & limits Model-specific | Default— |
Fieldmodalities | Typestring[] | RequirementOptional | Values & limits text | Default— |
Fieldprovider | Typeobject | RequirementOptional | Values & limits Conditions: provider.options is rejected on this Ultra text route | Default— |
Fieldcache_control.ttl | Typestring | RequirementOptional | Values & limits Conditions: 1h is rejected in actual prompt-cache locations | Default— |
Text · UTF8 text
messages[].content
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages |
Parametermaterial_types | Contracttext |
Parametermessages | Contractmessage array with role/content shape |
Parameterfile_material | Contractactual UTF-8 document inserted as literal text; no file ID inferred |
Parameteroptional_controls | Contractstream, temperature, top_p, stop, max_tokens, max_completion_tokens, reasoning, response_format, tools, tool_choice |
Parametercontrol_scope | Contractmodel-specific; user function tools differ from hosted tools |
Parameteroutput_modalities | Contracttext |
Parameterhosted_operations | Contractfalse |
Parametermultimodal_input_binding | Contractnull |
Lifecycle: synchronous_or_streamed
Send exact model ID with messages.
Context length is not an output cap.
A text file may supply literal text; image/audio/file-object inputs need exact binding.
Tariff references do not add request options. Supported parameters are listed for the selected method.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationRequest-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "~deepseek/deepseek-v4-flash-latest", "messages": [{"role": "user", "content": "Explain the difference between an observation and a hypothesis in three sentences."}]}')\n\nPath("request.json").write_text(json.dumps(p,ensure_ascii=False))\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/chat/completions \\\n -H "Authorization: Bearer $AILOOT_API_KEY" -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport json\nfrom pathlib import Path\nr=json.loads(Path("result.json").read_text())\nfor x in r.get("choices",[]):\n if x.get("message",{}).get("content"): print(x["message"]["content"])\nPY_RESULT\nfrom openai import OpenAI\nimport os\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nresult=client.chat.completions.create(model='~deepseek/deepseek-v4-flash-latest',messages=[{"role":"user","content":"Explain observations and hypotheses in three sentences."}])\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nsave_result(result.to_dict(mode="json"))\nfrom openai import OpenAI\nimport os\nfrom pathlib import Path\n\nif not os.environ.get("AILOOT_API_KEY", "").strip(): raise SystemExit("Set AILOOT_API_KEY before making a request")\nclient=OpenAI(base_url="https://api.ailoot.ai/v1",api_key=os.environ["AILOOT_API_KEY"],max_retries=0)\nimport base64, json\nfrom pathlib import Path\n\ndef save_result(data, stem="result"):\n # Keep every field, including usage, IDs, reasoning and tool requests.\n Path(stem + ".json").write_text(json.dumps(data, ensure_ascii=False, indent=2), encoding="utf-8")\n print(json.dumps(data, ensure_ascii=False, indent=2))\n def save_blob(encoded, suffix, label):\n raw = base64.b64decode(encoded, altchars=b"-_", validate=True)\n ext = "png" if raw.startswith(b"\\x89PNG\\r\\n\\x1a\\n") else "jpg" if raw.startswith(b"\\xff\\xd8") else "webp" if raw[:4] == b"RIFF" and raw[8:12] == b"WEBP" else suffix\n file = Path(stem + "-" + label + "." + ext)\n file.write_bytes(raw)\n print("Saved", file)\n for index, block in enumerate(data.get("content") or []):\n print("Content", index, json.dumps(block, ensure_ascii=False))\n for choice in data.get("choices") or []:\n print("Choice", choice.get("index"), json.dumps(choice.get("message", choice.get("delta", {})), ensure_ascii=False))\n for index, item in enumerate(data.get("output") or []):\n print("Output", index, json.dumps(item, ensure_ascii=False))\n if item.get("type") == "image_generation_call" and item.get("result"):\n save_blob(item["result"], item.get("output_format") or "bin", "output-" + str(index))\n for ci, candidate in enumerate(data.get("candidates") or []):\n for pi, part in enumerate((candidate.get("content") or {}).get("parts") or []):\n print("Candidate/part", ci, pi, json.dumps(part, ensure_ascii=False))\n blob = part.get("inlineData") or part.get("inline_data")\n if blob and blob.get("data"):\n mime = blob.get("mimeType") or blob.get("mime_type")\n suffix = {"image/png":"png", "image/jpeg":"jpg", "image/webp":"webp", "audio/wav":"wav", "audio/mpeg":"mp3", "video/mp4":"mp4"}.get(mime, "bin")\n save_blob(blob["data"], suffix, "candidate-" + str(ci) + "-part-" + str(pi))\n\nwith client.chat.completions.stream(model='~deepseek/deepseek-v4-flash-latest',messages=[{"role":"user","content":"Explain observations and hypotheses in three sentences."}]) as stream:\n with Path("stream-events.jsonl").open("w",encoding="utf-8") as log:\n for event in stream:\n payload=event.to_dict(mode="json")\n log.write(json.dumps(payload,ensure_ascii=False)+"\\n"); log.flush()\n print(json.dumps(payload,ensure_ascii=False))\n final=stream.get_final_completion()\n save_result(final.to_dict(mode="json"),"stream-result")\njson or SSE
Keep every complete returned response, file and content part.
Keep every choice and its content/tool calls/reasoning. In SSE assemble deltas by choice/tool index, retain the final finish reason and usage.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldchoices[].index | Typeinteger |
Fieldchoices[].message.content | Typestring|null |
Fieldchoices[].message.tool_calls | Typetool_call[] |
Fieldchoices[].message.reasoning_content | Typestring |
Fieldchoices[].message.reasoning_details | Typeobject|array |
Fieldchoices[].finish_reason | Typestring|null |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot deepseek-v4-flash-0731
Ailoot Ultra deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest
Use the method paths shown above.
deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest, deepseek-v4-flash-0731
~deepseek/deepseek-v4-flash-latestDeepSeek V4 Flash 0731 is the official Flash release for text, coding and agent workflows. It offers direct chat and thinking modes, structured tool-call output and a million-token model context; its dated edition includes the DSpark speculative-decoding module.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V4 Flash 0731 is the official release of the Flash branch, replacing its preview edition. Its practical focus includes coding and agent work: interpreting a supplied repository, proposing a change, using tool observations and returning an artifact the caller can inspect. Keep the 0731 edition attached to a saved result when comparing it with Flash Preview, Pro 0813 or V4.1 Flash.
Provide the actual task, the relevant files or document passages, the constraints and the expected result. For a repository change, include the affected code, the observed behavior and the command or request that reproduces the problem. Specify whether the deliverable is an explanation, a proposed patch, a list of findings or a structured plan.
A useful debugging brief is: “Use the supplied route handler, validation schema and failing request. Explain where the input diverges from the schema, propose the smallest change and identify the existing check that would distinguish a corrected result.” For a document task, name the supplied sections and ask for a table of requirements, evidence and unresolved questions. These are illustrative task briefs; include the real source material with the request.
The producer format supports direct chat and thinking. Direct chat returns response content directly. Thinking separates the model’s reasoning field from its final answer. In thinking mode, the documented effort choices are low, high and max, with low as the default. This edition uses those named levels, rather than V4.1 Flash’s integer effort scale.
Choose the mode around the work. A short wording change can specify the facts to retain and the tone of the replacement. A complicated repository task can define the investigation steps, the relevant evidence and a compact final report. Keep the desired final artifact explicit even when using a higher thinking effort: for example, a proposed diff with a short explanation and the commands needed to inspect it.
The native format supports multi-turn messages and tool definitions with named parameters and schemas. The model can return a requested tool invocation; the application executes the operation and supplies the resulting observation on the next turn. Keep tool results associated with their corresponding calls and preserve their order when returning multiple observations.
For a file-inspection workflow, expose the actual file-reading operation, describe its input path and return the observed contents. Ask the model to ground its next proposal in those contents. For a data workflow, provide the query operation and return the actual records; ask for a calculation or summary over those records. A returned call and an executed operation are separate steps in obtaining the requested result.
The producer’s encoder preserves the reasoning context across tool conversations. In ordinary chat, it normally removes reasoning from earlier turns before the latest user message. Use the documented message fields and conversation handling when retaining a multi-step workflow. The reference parser expects well-formed output, so check a parsed call’s name and parameters before executing it and retain an invalid result for inspection.
The dated model configuration specifies 1,048,576 sequence positions. Include the supplied material, conversation history, tool definitions, returned observations and generated response when preparing a sequence. Give the model a useful answer target, such as a small patch, an evidence table or a numbered plan, rather than filling the available space with unrelated files.
The maker’s local recipe recommends temperature 1.0, with top_p 0.95 for agent work and 1.0 otherwise. For high and max thinking effort it recommends setting a maximum output length of 384K tokens. Those are published recipe settings and a generation-budget recommendation; the sequence budget still has to accommodate the rest of the request.
The 0731 release uses the Flash DSpark structure, including an attached speculative-decoding module. The underlying Flash backbone is a mixture-of-experts model described with 284 billion total parameters and 13 billion activated per token. Keep these backbone counts distinct from any additional module weights. The exact dated configuration has 43 layers, 256 routed experts, one shared expert and six selected experts per token.
The useful output remains the answer, proposed code, structured finding or next workflow step requested by the caller. Inspect a code change against the supplied files and observed failure. Inspect a document conclusion against the cited passage or returned record. Save the model edition, task brief and relevant observations with the result so a later comparison uses the same problem and evidence.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits ~deepseek/deepseek-v4-flash-latest | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits user, assistant | Default— |
Fieldmessages[].content | Typestring | RequirementRequired | Values & limits | Default— |
Fieldmax_tokens | Typeinteger | RequirementRequired | Values & limits | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
Text · UTF-8 text
messages[].content string
| Parameter | Contract |
|---|---|
Parametercontent_type | Contractapplication/json |
Parameterrequired | Contractmodel, messages, max_tokens |
Parameteroptional_controls | Contractstream |
Parametermaterial_types | Contracttext |
Parametertext_file_material | ContractRead the actual UTF-8 document and put its text in the indicated input or message content string. |
Lifecycle: synchronous_or_streamed
Tariff references do not add request options. Supported parameters are listed for the selected method.
Match each charge to the conditions shown. Use one rate for each set of alternative configurations.
| Configuration | Rate schedule | Context tokens, more than | Context tokens, up to and including | Charge |
|---|---|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Context tokens, more than— | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
| ConfigurationRequest-length tier 1 | Rate schedule— | Context tokens, more than0 | Context tokens, up to and includingNo upper bound | Charge Input · Input tokens $0.00088 / 1K tokens Output · Output tokens $0.00264 / 1K tokens Input · Cache read $0.000056 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "~deepseek/deepseek-v4-flash-latest",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\n: "${AILOOT_TEXT_FILE:?Set your actual UTF-8 document}"\nexport AILOOT_TEXT_FILE\npython3 - <<'PY_REQUEST'\nimport json, os\nfrom pathlib import Path\n\nsource_file = Path(os.environ['AILOOT_TEXT_FILE'])\nif not source_file.is_file():\n raise SystemExit('Set AILOOT_TEXT_FILE to your actual UTF-8 document')\ntext = source_file.read_text(encoding='utf-8')\nif not text.strip():\n raise SystemExit('Choose a document containing text')\npayload = {'model': '~deepseek/deepseek-v4-flash-latest', 'max_tokens': 128, 'messages': [{'role': 'user', 'content': 'Summarize the incident report.'}]}\npayload['messages'][0]['content'] = 'Summarize this material and quote supporting source passages.\\n\\n' + text\nPath('request.json').write_text(json.dumps(payload, ensure_ascii=False), encoding='utf-8')\nPY_REQUEST\ncurl --fail-with-body --silent --show-error https://api.ailoot.ai/v1/messages \\\n -H "x-api-key: $AILOOT_API_KEY" -H 'anthropic-version: 2023-06-01' -H 'Content-Type: application/json' \\\n --data-binary @request.json -D response-headers.txt -o result.json\npython3 - <<'PY_RESULT'\nimport json\nfrom pathlib import Path\n\nresult = json.loads(Path('result.json').read_text(encoding='utf-8'))\nprint(json.dumps(result, ensure_ascii=False, indent=2))\nPY_RESULT\nimport json, os\nfrom pathlib import Path\n\nsource_file = Path(os.environ['AILOOT_TEXT_FILE'])\nif not source_file.is_file():\n raise SystemExit('Set AILOOT_TEXT_FILE to your actual UTF-8 document')\ntext = source_file.read_text(encoding='utf-8')\nif not text.strip():\n raise SystemExit('Choose a document containing text')\npayload = {'model': '~deepseek/deepseek-v4-flash-latest', 'max_tokens': 128, 'messages': [{'role': 'user', 'content': 'Summarize the incident report.'}]}\npayload['messages'][0]['content'] = 'Summarize this material and quote supporting source passages.\\n\\n' + text\nfrom urllib.request import Request, urlopen\nfrom urllib.error import HTTPError\n\napi_key = os.environ['AILOOT_API_KEY']\nif not api_key.strip():\n raise SystemExit('Set AILOOT_API_KEY before making a request')\nheaders = {'x-api-key': 'PLACEHOLDER_KEY', 'anthropic-version': '2023-06-01', 'Content-Type': 'application/json'}\nheaders['x-api-key'] = api_key\nrequest = Request('https://api.ailoot.ai/v1/messages', data=json.dumps(payload, ensure_ascii=False).encode('utf-8'), headers=headers, method='POST')\ntry:\n with urlopen(request, timeout=120) as response:\n raw = response.read()\n Path('response-headers.txt').write_text(str(response.headers), encoding='utf-8')\nexcept HTTPError as error:\n Path('result.json').write_bytes(error.read())\n Path('response-headers.txt').write_text(str(error.headers), encoding='utf-8')\n raise\nPath('result.json').write_bytes(raw)\nimport json\nfrom pathlib import Path\n\nresult = json.loads(Path('result.json').read_text(encoding='utf-8'))\nprint(json.dumps(result, ensure_ascii=False, indent=2))\nJSON or SSE
Keep every complete returned response, file and content part.
Keep all content blocks, thinking signatures and actual tool-use IDs. Assemble all SSE block deltas; a first text block is not the complete output.
Keep request/response IDs and headers. No generic text job retrieval URL is supplied.
| Field | Type |
|---|---|
Fieldid | Typestring |
Fieldmodel | Typestring |
Fieldcontent[] | Typetext|thinking|tool_use|other content block |
Fieldstop_reason | Typestring|null |
Fieldusage | Typeobject |
| Response | Condition | Next step |
|---|---|---|
Response400 · invalid_request_error | ConditionInvalid required fields, material representation or parameter value | Next stepCorrect the indicated fields before a new creation request. |
Ailoot deepseek-v4-flash-0731
Ailoot Ultra deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest
Use the method paths shown above.
deepseek/deepseek-v4-flash-0731, ~deepseek/deepseek-v4-flash-latest, deepseek-v4-flash-0731
Technical example in English
Repair a required-field check\n\nA route rejects only if not payload.get("quantity"). The contract requires quantity to be an integer greater than zero, excluding booleans. Explain why inputs {"quantity": -2} and {"quantity": true} expose a problem. Propose a small Python validation patch and cases for missing, zero, negative, boolean and positive values.A diagnosis, a focused validation patch and five input cases with accept or reject expectations.Check that Python booleans are not accepted as integers and positive whole quantities still pass.