DeepSeek V3 is DeepSeek’s original mixture-of-experts text model for general language, coding and reasoning work. It takes written instructions and text context and returns an answer, draft, explanation or code. The model card lists 671 billion total main-model parameters with 37 billion active for each token. The named original V3 remains separate from V3-0324, V3.1 and V3.2, whose updates add their own features and prompting profiles.
Its architecture combines Multi-head Latent Attention with DeepSeekMoE. Latent attention compresses information used by attention, while sparse expert selection activates only part of the model for each token. V3 also introduces auxiliary-loss-free load balancing and a multi-token prediction training objective. The published model package includes a separate 14-billion-parameter multi-token prediction module in addition to the 671-billion-parameter main model, explaining why package totals can differ from the main-model architecture count.
DeepSeek pre-trains V3 on 14.8 trillion tokens and then uses supervised fine-tuning and reinforcement learning. The post-training process brings verification and reflection patterns from the R1 family into a standard language-model response while controlling output style and length. This is a training contribution, distinct from choosing R1 itself or the later V3.1 hybrid thinking template. Prepare the final answer format you need rather than assuming one reasoning presentation from another edition.
The producer model table gives a 128K context for both V3 Base and the post-trained V3 edition. That context can join a question with substantial document or code material, while conversation history and the response also need space. Label the sources and identify the relevant question. A context window is a capacity bound, not an instruction to submit unrelated material or ask for a needlessly long answer.
For a practical code task, supply the relevant function, expected behaviour, important constraints and the input that demonstrates the problem. Ask for a proposed correction and an explanation tied to the supplied code. For document analysis, provide named sections and clear comparison criteria. An illustrative brief is: “Extract the requirements, identify which are addressed by this proposal and cite the corresponding passage for every finding. Finish with a concise list of missing details.”
Writing and editing tasks benefit from an explicit audience, tone and set of facts that must be retained. Ask for a summary, rewrite, translation or explanation and state the mandatory names and terminology. A requested list of fields or a JSON-shaped response is still something to review as generated content; the model name alone does not qualify a surrounding application’s structured-output or tool-execution mechanism. If an application supplies tools, keep each returned observation distinct from the action the model proposed.
The manufacturer publishes FP8 model material and describes other precision or execution options in its reference. Those implementation choices are separate from the source-level task and should not be inferred from the displayed model name. The model’s output can be code, but generating that code does not run it; an explanation of a source does not prove that an external lookup was performed. Use actual source passages and observed application results when the answer needs that evidence.
For a useful result, define the final artifact and the conditions under which it is complete. Keep source-supported findings, assumptions and proposed changes separate, then review the generated answer against the supplied material. A longer conversation can refine one part of the draft while preserving approved facts and constraints. The deliverable is a reviewable text or code artifact for the stated task, with the original V3 identity retained rather than silently borrowing a later edition’s controls.
", "aboutText": "DeepSeek V3 is DeepSeek’s original mixture-of-experts text model for general language, coding and reasoning work. It takes written instructions and text context and returns an answer, draft, explanation or code. The model card lists 671 billion total main-model parameters with 37 billion active for each token. The named original V3 remains separate from V3-0324, V3.1 and V3.2, whose updates add their own features and prompting profiles.\n\nIts architecture combines Multi-head Latent Attention with DeepSeekMoE. Latent attention compresses information used by attention, while sparse expert selection activates only part of the model for each token. V3 also introduces auxiliary-loss-free load balancing and a multi-token prediction training objective. The published model package includes a separate 14-billion-parameter multi-token prediction module in addition to the 671-billion-parameter main model, explaining why package totals can differ from the main-model architecture count.\n\nDeepSeek pre-trains V3 on 14.8 trillion tokens and then uses supervised fine-tuning and reinforcement learning. The post-training process brings verification and reflection patterns from the R1 family into a standard language-model response while controlling output style and length. This is a training contribution, distinct from choosing R1 itself or the later V3.1 hybrid thinking template. Prepare the final answer format you need rather than assuming one reasoning presentation from another edition.\n\nThe producer model table gives a 128K context for both V3 Base and the post-trained V3 edition. That context can join a question with substantial document or code material, while conversation history and the response also need space. Label the sources and identify the relevant question. A context window is a capacity bound, not an instruction to submit unrelated material or ask for a needlessly long answer.\n\nFor a practical code task, supply the relevant function, expected behaviour, important constraints and the input that demonstrates the problem. Ask for a proposed correction and an explanation tied to the supplied code. For document analysis, provide named sections and clear comparison criteria. An illustrative brief is: “Extract the requirements, identify which are addressed by this proposal and cite the corresponding passage for every finding. Finish with a concise list of missing details.”\n\nWriting and editing tasks benefit from an explicit audience, tone and set of facts that must be retained. Ask for a summary, rewrite, translation or explanation and state the mandatory names and terminology. A requested list of fields or a JSON-shaped response is still something to review as generated content; the model name alone does not qualify a surrounding application’s structured-output or tool-execution mechanism. If an application supplies tools, keep each returned observation distinct from the action the model proposed.\n\nThe manufacturer publishes FP8 model material and describes other precision or execution options in its reference. Those implementation choices are separate from the source-level task and should not be inferred from the displayed model name. The model’s output can be code, but generating that code does not run it; an explanation of a source does not prove that an external lookup was performed. Use actual source passages and observed application results when the answer needs that evidence.\n\nFor a useful result, define the final artifact and the conditions under which it is complete. Keep source-supported findings, assumptions and proposed changes separate, then review the generated answer against the supplied material. A longer conversation can refine one part of the draft while preserving approved facts and constraints. The deliverable is a reviewable text or code artifact for the stated task, with the original V3 identity retained rather than silently borrowing a later edition’s controls.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "d9f3585d31eca8bb27e0e98926f1675ae919e2c49a69016ddbd3c7084cbde59c", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "deepseek-chatDeepSeek V3 is DeepSeek’s original text model for coding, writing, questions and source-based analysis. Its mixture-of-experts architecture has 671B total and 37B active parameters, with a 128K context and post-training that incorporates reasoning patterns from R1.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V3 is DeepSeek’s original mixture-of-experts text model for general language, coding and reasoning work. It takes written instructions and text context and returns an answer, draft, explanation or code. The model card lists 671 billion total main-model parameters with 37 billion active for each token. The named original V3 remains separate from V3-0324, V3.1 and V3.2, whose updates add their own features and prompting profiles.
Its architecture combines Multi-head Latent Attention with DeepSeekMoE. Latent attention compresses information used by attention, while sparse expert selection activates only part of the model for each token. V3 also introduces auxiliary-loss-free load balancing and a multi-token prediction training objective. The published model package includes a separate 14-billion-parameter multi-token prediction module in addition to the 671-billion-parameter main model, explaining why package totals can differ from the main-model architecture count.
DeepSeek pre-trains V3 on 14.8 trillion tokens and then uses supervised fine-tuning and reinforcement learning. The post-training process brings verification and reflection patterns from the R1 family into a standard language-model response while controlling output style and length. This is a training contribution, distinct from choosing R1 itself or the later V3.1 hybrid thinking template. Prepare the final answer format you need rather than assuming one reasoning presentation from another edition.
The producer model table gives a 128K context for both V3 Base and the post-trained V3 edition. That context can join a question with substantial document or code material, while conversation history and the response also need space. Label the sources and identify the relevant question. A context window is a capacity bound, not an instruction to submit unrelated material or ask for a needlessly long answer.
For a practical code task, supply the relevant function, expected behaviour, important constraints and the input that demonstrates the problem. Ask for a proposed correction and an explanation tied to the supplied code. For document analysis, provide named sections and clear comparison criteria. An illustrative brief is: “Extract the requirements, identify which are addressed by this proposal and cite the corresponding passage for every finding. Finish with a concise list of missing details.”
Writing and editing tasks benefit from an explicit audience, tone and set of facts that must be retained. Ask for a summary, rewrite, translation or explanation and state the mandatory names and terminology. A requested list of fields or a JSON-shaped response is still something to review as generated content; the model name alone does not qualify a surrounding application’s structured-output or tool-execution mechanism. If an application supplies tools, keep each returned observation distinct from the action the model proposed.
The manufacturer publishes FP8 model material and describes other precision or execution options in its reference. Those implementation choices are separate from the source-level task and should not be inferred from the displayed model name. The model’s output can be code, but generating that code does not run it; an explanation of a source does not prove that an external lookup was performed. Use actual source passages and observed application results when the answer needs that evidence.
For a useful result, define the final artifact and the conditions under which it is complete. Keep source-supported findings, assumptions and proposed changes separate, then review the generated answer against the supplied material. A longer conversation can refine one part of the draft while preserving approved facts and constraints. The deliverable is a reviewable text or code artifact for the stated task, with the original V3 identity retained rather than silently borrowing a later edition’s controls.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek-chat | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits Minimum items: 1 | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits Model-specific | Default— |
Fieldmessages[].content | Typestring | RequirementOptional | Values & limits Applies when: Text conversation | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
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 |
Parametercontrol_scope | ContractText conversation; stream is a boolean and defaults to false. |
Parameteroutput_modalities | Contracttext |
Parameterhosted_operations | Contractfalse |
Parametermultimodal_input_binding | Contractnull |
Method aliases: /chat/completions
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.
| Configuration | Rate schedule | Charge |
|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Charge Input · Input tokens $0.00064 / 1K tokens Output · Output tokens $0.00178 / 1K tokens Input · Cache read $0 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model":"deepseek-chat","messages":[{"role":"user","content":"hello"}],"stream":false}')\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-chat", messages=[{"role":"user","content":"hello"}], stream=False)\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"))\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 Ultra deepseek-chat, deepseek/deepseek-chat
/chat/completions
deepseek-chatDeepSeek V3 is DeepSeek’s original text model for coding, writing, questions and source-based analysis. Its mixture-of-experts architecture has 671B total and 37B active parameters, with a 128K context and post-training that incorporates reasoning patterns from R1.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V3 is DeepSeek’s original mixture-of-experts text model for general language, coding and reasoning work. It takes written instructions and text context and returns an answer, draft, explanation or code. The model card lists 671 billion total main-model parameters with 37 billion active for each token. The named original V3 remains separate from V3-0324, V3.1 and V3.2, whose updates add their own features and prompting profiles.
Its architecture combines Multi-head Latent Attention with DeepSeekMoE. Latent attention compresses information used by attention, while sparse expert selection activates only part of the model for each token. V3 also introduces auxiliary-loss-free load balancing and a multi-token prediction training objective. The published model package includes a separate 14-billion-parameter multi-token prediction module in addition to the 671-billion-parameter main model, explaining why package totals can differ from the main-model architecture count.
DeepSeek pre-trains V3 on 14.8 trillion tokens and then uses supervised fine-tuning and reinforcement learning. The post-training process brings verification and reflection patterns from the R1 family into a standard language-model response while controlling output style and length. This is a training contribution, distinct from choosing R1 itself or the later V3.1 hybrid thinking template. Prepare the final answer format you need rather than assuming one reasoning presentation from another edition.
The producer model table gives a 128K context for both V3 Base and the post-trained V3 edition. That context can join a question with substantial document or code material, while conversation history and the response also need space. Label the sources and identify the relevant question. A context window is a capacity bound, not an instruction to submit unrelated material or ask for a needlessly long answer.
For a practical code task, supply the relevant function, expected behaviour, important constraints and the input that demonstrates the problem. Ask for a proposed correction and an explanation tied to the supplied code. For document analysis, provide named sections and clear comparison criteria. An illustrative brief is: “Extract the requirements, identify which are addressed by this proposal and cite the corresponding passage for every finding. Finish with a concise list of missing details.”
Writing and editing tasks benefit from an explicit audience, tone and set of facts that must be retained. Ask for a summary, rewrite, translation or explanation and state the mandatory names and terminology. A requested list of fields or a JSON-shaped response is still something to review as generated content; the model name alone does not qualify a surrounding application’s structured-output or tool-execution mechanism. If an application supplies tools, keep each returned observation distinct from the action the model proposed.
The manufacturer publishes FP8 model material and describes other precision or execution options in its reference. Those implementation choices are separate from the source-level task and should not be inferred from the displayed model name. The model’s output can be code, but generating that code does not run it; an explanation of a source does not prove that an external lookup was performed. Use actual source passages and observed application results when the answer needs that evidence.
For a useful result, define the final artifact and the conditions under which it is complete. Keep source-supported findings, assumptions and proposed changes separate, then review the generated answer against the supplied material. A longer conversation can refine one part of the draft while preserving approved facts and constraints. The deliverable is a reviewable text or code artifact for the stated task, with the original V3 identity retained rather than silently borrowing a later edition’s controls.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek-chat | 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.
| Configuration | Rate schedule | Charge |
|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Charge Input · Input tokens $0.00064 / 1K tokens Output · Output tokens $0.00178 / 1K tokens Input · Cache read $0 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "deepseek-chat",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${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-chat', '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-chat', '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 Ultra deepseek-chat, deepseek/deepseek-chat
/chat/completions
deepseek/deepseek-chatDeepSeek V3 is DeepSeek’s original text model for coding, writing, questions and source-based analysis. Its mixture-of-experts architecture has 671B total and 37B active parameters, with a 128K context and post-training that incorporates reasoning patterns from R1.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V3 is DeepSeek’s original mixture-of-experts text model for general language, coding and reasoning work. It takes written instructions and text context and returns an answer, draft, explanation or code. The model card lists 671 billion total main-model parameters with 37 billion active for each token. The named original V3 remains separate from V3-0324, V3.1 and V3.2, whose updates add their own features and prompting profiles.
Its architecture combines Multi-head Latent Attention with DeepSeekMoE. Latent attention compresses information used by attention, while sparse expert selection activates only part of the model for each token. V3 also introduces auxiliary-loss-free load balancing and a multi-token prediction training objective. The published model package includes a separate 14-billion-parameter multi-token prediction module in addition to the 671-billion-parameter main model, explaining why package totals can differ from the main-model architecture count.
DeepSeek pre-trains V3 on 14.8 trillion tokens and then uses supervised fine-tuning and reinforcement learning. The post-training process brings verification and reflection patterns from the R1 family into a standard language-model response while controlling output style and length. This is a training contribution, distinct from choosing R1 itself or the later V3.1 hybrid thinking template. Prepare the final answer format you need rather than assuming one reasoning presentation from another edition.
The producer model table gives a 128K context for both V3 Base and the post-trained V3 edition. That context can join a question with substantial document or code material, while conversation history and the response also need space. Label the sources and identify the relevant question. A context window is a capacity bound, not an instruction to submit unrelated material or ask for a needlessly long answer.
For a practical code task, supply the relevant function, expected behaviour, important constraints and the input that demonstrates the problem. Ask for a proposed correction and an explanation tied to the supplied code. For document analysis, provide named sections and clear comparison criteria. An illustrative brief is: “Extract the requirements, identify which are addressed by this proposal and cite the corresponding passage for every finding. Finish with a concise list of missing details.”
Writing and editing tasks benefit from an explicit audience, tone and set of facts that must be retained. Ask for a summary, rewrite, translation or explanation and state the mandatory names and terminology. A requested list of fields or a JSON-shaped response is still something to review as generated content; the model name alone does not qualify a surrounding application’s structured-output or tool-execution mechanism. If an application supplies tools, keep each returned observation distinct from the action the model proposed.
The manufacturer publishes FP8 model material and describes other precision or execution options in its reference. Those implementation choices are separate from the source-level task and should not be inferred from the displayed model name. The model’s output can be code, but generating that code does not run it; an explanation of a source does not prove that an external lookup was performed. Use actual source passages and observed application results when the answer needs that evidence.
For a useful result, define the final artifact and the conditions under which it is complete. Keep source-supported findings, assumptions and proposed changes separate, then review the generated answer against the supplied material. A longer conversation can refine one part of the draft while preserving approved facts and constraints. The deliverable is a reviewable text or code artifact for the stated task, with the original V3 identity retained rather than silently borrowing a later edition’s controls.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek/deepseek-chat | Default— |
Fieldmessages | Typemessage[] | RequirementRequired | Values & limits Minimum items: 1 | Default— |
Fieldmessages[].role | Typestring | RequirementRequired | Values & limits Model-specific | Default— |
Fieldmessages[].content | Typestring | RequirementOptional | Values & limits Applies when: Text conversation | Default— |
Fieldstream | Typeboolean | RequirementOptional | Values & limits | Defaultfalse |
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 |
Parametercontrol_scope | ContractText conversation; stream is a boolean and defaults to false. |
Parameteroutput_modalities | Contracttext |
Parameterhosted_operations | Contractfalse |
Parametermultimodal_input_binding | Contractnull |
Method aliases: /chat/completions
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.
| Configuration | Rate schedule | Charge |
|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Charge Input · Input tokens $0.00064 / 1K tokens Output · Output tokens $0.00178 / 1K tokens Input · Cache read $0 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model":"deepseek/deepseek-chat","messages":[{"role":"user","content":"hello"}],"stream":false}')\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-chat", messages=[{"role":"user","content":"hello"}], stream=False)\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"))\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 Ultra deepseek-chat, deepseek/deepseek-chat
/chat/completions
deepseek/deepseek-chatDeepSeek V3 is DeepSeek’s original text model for coding, writing, questions and source-based analysis. Its mixture-of-experts architecture has 671B total and 37B active parameters, with a 128K context and post-training that incorporates reasoning patterns from R1.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
DeepSeek V3 is DeepSeek’s original mixture-of-experts text model for general language, coding and reasoning work. It takes written instructions and text context and returns an answer, draft, explanation or code. The model card lists 671 billion total main-model parameters with 37 billion active for each token. The named original V3 remains separate from V3-0324, V3.1 and V3.2, whose updates add their own features and prompting profiles.
Its architecture combines Multi-head Latent Attention with DeepSeekMoE. Latent attention compresses information used by attention, while sparse expert selection activates only part of the model for each token. V3 also introduces auxiliary-loss-free load balancing and a multi-token prediction training objective. The published model package includes a separate 14-billion-parameter multi-token prediction module in addition to the 671-billion-parameter main model, explaining why package totals can differ from the main-model architecture count.
DeepSeek pre-trains V3 on 14.8 trillion tokens and then uses supervised fine-tuning and reinforcement learning. The post-training process brings verification and reflection patterns from the R1 family into a standard language-model response while controlling output style and length. This is a training contribution, distinct from choosing R1 itself or the later V3.1 hybrid thinking template. Prepare the final answer format you need rather than assuming one reasoning presentation from another edition.
The producer model table gives a 128K context for both V3 Base and the post-trained V3 edition. That context can join a question with substantial document or code material, while conversation history and the response also need space. Label the sources and identify the relevant question. A context window is a capacity bound, not an instruction to submit unrelated material or ask for a needlessly long answer.
For a practical code task, supply the relevant function, expected behaviour, important constraints and the input that demonstrates the problem. Ask for a proposed correction and an explanation tied to the supplied code. For document analysis, provide named sections and clear comparison criteria. An illustrative brief is: “Extract the requirements, identify which are addressed by this proposal and cite the corresponding passage for every finding. Finish with a concise list of missing details.”
Writing and editing tasks benefit from an explicit audience, tone and set of facts that must be retained. Ask for a summary, rewrite, translation or explanation and state the mandatory names and terminology. A requested list of fields or a JSON-shaped response is still something to review as generated content; the model name alone does not qualify a surrounding application’s structured-output or tool-execution mechanism. If an application supplies tools, keep each returned observation distinct from the action the model proposed.
The manufacturer publishes FP8 model material and describes other precision or execution options in its reference. Those implementation choices are separate from the source-level task and should not be inferred from the displayed model name. The model’s output can be code, but generating that code does not run it; an explanation of a source does not prove that an external lookup was performed. Use actual source passages and observed application results when the answer needs that evidence.
For a useful result, define the final artifact and the conditions under which it is complete. Keep source-supported findings, assumptions and proposed changes separate, then review the generated answer against the supplied material. A longer conversation can refine one part of the draft while preserving approved facts and constraints. The deliverable is a reviewable text or code artifact for the stated task, with the original V3 identity retained rather than silently borrowing a later edition’s controls.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits deepseek/deepseek-chat | 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.
| Configuration | Rate schedule | Charge |
|---|---|---|
| ConfigurationBase token rates | Rate scheduleBase rate | Charge Input · Input tokens $0.00064 / 1K tokens Output · Output tokens $0.00178 / 1K tokens Input · Cache read $0 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "deepseek/deepseek-chat",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${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-chat', '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-chat', '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 Ultra deepseek-chat, deepseek/deepseek-chat
/chat/completions