Qwen3.6 highlights: coding and reasoning continuity
Qwen3.6-35B-A3B is a vision-language model focused on useful coding and agent work. Qwen highlights stronger frontend workflows and repository-level reasoning, plus a new option to retain thinking from historical messages. It accepts text, images and video and returns text: an explanation, code proposal, structured analysis or another requested textual artifact. Preserve this exact 35B-A3B edition rather than treating it as the dense Qwen3.6-27B or its earlier Qwen3.5 counterpart.
For a frontend task, provide the relevant screenshot, functional requirements and existing code context. Ask for editable source and a description of the intended states and interactions. For a repository task, identify the affected feature, the input exposing a problem and expected behavior. Request a proposed patch with its assumptions and relevant checks.
Model overview
The mixture-of-experts model has 35 billion parameters in total and activates three billion. Its 40-layer hybrid layout repeats three Gated DeltaNet expert layers followed by one gated-attention expert layer. It has 256 routed experts and selects eight routed experts with one shared expert; multi-token prediction is part of training. The smaller activated count does not describe the full checkpoint size. A vision encoder supplies the visual part of the model.
The native context is 262,144 tokens, with conditional extension to 1,010,000. The maker recommends retaining a context of at least 128K when possible for complex thinking tasks, even if reducing it would ease memory pressure. Source text, represented visuals, historical messages, reasoning and the final answer share the configured sequence allowance.
Using inputs and response modes
The card shows text, image and video questions. For a diagram or page, preserve the labels and define the exact question. For a clip, keep timing and state the event of interest. The maker’s supported video example defaults to two frames per second and allows sampling configuration. The best-practice recipe expands the video processor’s budget to roughly 224K video tokens for higher-rate hour-scale analysis; it is a configuration example rather than a hard universal clip-duration limit.
Thinking is enabled by default. The maker’s enable_thinking control can disable it for a direct answer. Qwen3.6 does not officially support the older Qwen3 soft prompt switches /think and /nothink. Separate the reasoning from the final artifact when presenting a result, and follow this checkpoint’s own interface rather than assume prompt markers or controls from another generation will apply.
Preserving historical thinking
By default, the template retains thinking blocks associated with the latest user message. Qwen3.6 has additionally been trained to use historical reasoning when preserve_thinking is enabled. Its explicit example sets that option to true. This can help an iterative agent task retain decision context instead of repeatedly re-deriving earlier choices; it also changes the amount of history the request carries. The availability of the option depends on the interface exposing the maker’s recipe.
Agentic usage
Qwen documents tool calling and recommends Qwen-Agent for configuring tools, including MCP-connected tools. Qwen Code is another maker-documented coding-agent workflow. Define the permitted tools, task goal and expected deliverable, then return actual tool observations to the model. The model can reason about a result and revise a plan; access to files or execution environments comes from the connected application, not the model name.
Ultra-long text
The documented context extension uses YaRN with the checkpoint’s interleaved positional settings. A factor of four targets the approximately one-million-token workflow; for 524,288-token workloads, Qwen suggests a factor of two. Static scaling can affect shorter-text performance, so the maker recommends changing it when the longer context is needed. Retain section and code identifiers in a large input and ask the result to show which evidence supports each conclusion.
Generation guidance and limits
For general thinking, the maker recommends temperature 1.0, top_p 0.95, top_k 20, min_p 0, presence penalty 1.5 and repetition penalty 1.0. For precise coding, temperature becomes 0.6 and presence penalty zero. For direct responses, it recommends temperature 0.7, top_p 0.8 and presence penalty 1.5, with the other values unchanged. Higher presence penalties can reduce repetition but may mix languages or slightly reduce quality.
Qwen recommends a 32,768-token generation allowance for most questions and 81,920 for unusually complex mathematical or programming tasks. These are recommendations, not separately established hard output ceilings. State the final format explicitly and check code, calculations and extracted visual details against the source. Reasoning continuity and long context help carry a task’s evidence, while a checkable final artifact makes the result useful.
", "aboutText": "Qwen3.6 highlights: coding and reasoning continuity\n\nQwen3.6-35B-A3B is a vision-language model focused on useful coding and agent work. Qwen highlights stronger frontend workflows and repository-level reasoning, plus a new option to retain thinking from historical messages. It accepts text, images and video and returns text: an explanation, code proposal, structured analysis or another requested textual artifact. Preserve this exact 35B-A3B edition rather than treating it as the dense Qwen3.6-27B or its earlier Qwen3.5 counterpart.\n\nFor a frontend task, provide the relevant screenshot, functional requirements and existing code context. Ask for editable source and a description of the intended states and interactions. For a repository task, identify the affected feature, the input exposing a problem and expected behavior. Request a proposed patch with its assumptions and relevant checks.\n\nModel overview\n\nThe mixture-of-experts model has 35 billion parameters in total and activates three billion. Its 40-layer hybrid layout repeats three Gated DeltaNet expert layers followed by one gated-attention expert layer. It has 256 routed experts and selects eight routed experts with one shared expert; multi-token prediction is part of training. The smaller activated count does not describe the full checkpoint size. A vision encoder supplies the visual part of the model.\n\nThe native context is 262,144 tokens, with conditional extension to 1,010,000. The maker recommends retaining a context of at least 128K when possible for complex thinking tasks, even if reducing it would ease memory pressure. Source text, represented visuals, historical messages, reasoning and the final answer share the configured sequence allowance.\n\nUsing inputs and response modes\n\nThe card shows text, image and video questions. For a diagram or page, preserve the labels and define the exact question. For a clip, keep timing and state the event of interest. The maker’s supported video example defaults to two frames per second and allows sampling configuration. The best-practice recipe expands the video processor’s budget to roughly 224K video tokens for higher-rate hour-scale analysis; it is a configuration example rather than a hard universal clip-duration limit.\n\nThinking is enabled by default. The maker’s enable_thinking control can disable it for a direct answer. Qwen3.6 does not officially support the older Qwen3 soft prompt switches /think and /nothink. Separate the reasoning from the final artifact when presenting a result, and follow this checkpoint’s own interface rather than assume prompt markers or controls from another generation will apply.\n\nPreserving historical thinking\n\nBy default, the template retains thinking blocks associated with the latest user message. Qwen3.6 has additionally been trained to use historical reasoning when preserve_thinking is enabled. Its explicit example sets that option to true. This can help an iterative agent task retain decision context instead of repeatedly re-deriving earlier choices; it also changes the amount of history the request carries. The availability of the option depends on the interface exposing the maker’s recipe.\n\nAgentic usage\n\nQwen documents tool calling and recommends Qwen-Agent for configuring tools, including MCP-connected tools. Qwen Code is another maker-documented coding-agent workflow. Define the permitted tools, task goal and expected deliverable, then return actual tool observations to the model. The model can reason about a result and revise a plan; access to files or execution environments comes from the connected application, not the model name.\n\nUltra-long text\n\nThe documented context extension uses YaRN with the checkpoint’s interleaved positional settings. A factor of four targets the approximately one-million-token workflow; for 524,288-token workloads, Qwen suggests a factor of two. Static scaling can affect shorter-text performance, so the maker recommends changing it when the longer context is needed. Retain section and code identifiers in a large input and ask the result to show which evidence supports each conclusion.\n\nGeneration guidance and limits\n\nFor general thinking, the maker recommends temperature 1.0, top_p 0.95, top_k 20, min_p 0, presence penalty 1.5 and repetition penalty 1.0. For precise coding, temperature becomes 0.6 and presence penalty zero. For direct responses, it recommends temperature 0.7, top_p 0.8 and presence penalty 1.5, with the other values unchanged. Higher presence penalties can reduce repetition but may mix languages or slightly reduce quality.\n\nQwen recommends a 32,768-token generation allowance for most questions and 81,920 for unusually complex mathematical or programming tasks. These are recommendations, not separately established hard output ceilings. State the final format explicitly and check code, calculations and extracted visual details against the source. Reasoning continuity and long context help carry a task’s evidence, while a checkable final artifact makes the result useful.", "editorialOverlay": { "copy_fields": { "practical_html": "Technical example in English
Correct a button loading state\n\nPropose a small JavaScript change for a save form. Current handler sets button.disabled = true, awaits saveDraft(), then reenables the button. When saveDraft rejects, the button stays disabled. Preserve the save operation and show pseudocode for displaying an error and always restoring the button. Add success and failure acceptance checks.A focused async handler or pseudocode plus two checks covering button and error states.Verify the cleanup happens on rejection and success without converting a failed save into a success message.qwen/qwen3.6-35b-a3bQwen3.6-35B-A3B is Qwen’s vision-language mixture-of-experts model with 35 billion total and 3 billion activated parameters. It focuses on frontend and repository-level coding, adds optional preservation of historical reasoning, and accepts text, images and video with textual output.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Qwen3.6 highlights: coding and reasoning continuity
Qwen3.6-35B-A3B is a vision-language model focused on useful coding and agent work. Qwen highlights stronger frontend workflows and repository-level reasoning, plus a new option to retain thinking from historical messages. It accepts text, images and video and returns text: an explanation, code proposal, structured analysis or another requested textual artifact. Preserve this exact 35B-A3B edition rather than treating it as the dense Qwen3.6-27B or its earlier Qwen3.5 counterpart.
For a frontend task, provide the relevant screenshot, functional requirements and existing code context. Ask for editable source and a description of the intended states and interactions. For a repository task, identify the affected feature, the input exposing a problem and expected behavior. Request a proposed patch with its assumptions and relevant checks.
Model overview
The mixture-of-experts model has 35 billion parameters in total and activates three billion. Its 40-layer hybrid layout repeats three Gated DeltaNet expert layers followed by one gated-attention expert layer. It has 256 routed experts and selects eight routed experts with one shared expert; multi-token prediction is part of training. The smaller activated count does not describe the full checkpoint size. A vision encoder supplies the visual part of the model.
The native context is 262,144 tokens, with conditional extension to 1,010,000. The maker recommends retaining a context of at least 128K when possible for complex thinking tasks, even if reducing it would ease memory pressure. Source text, represented visuals, historical messages, reasoning and the final answer share the configured sequence allowance.
Using inputs and response modes
The card shows text, image and video questions. For a diagram or page, preserve the labels and define the exact question. For a clip, keep timing and state the event of interest. The maker’s supported video example defaults to two frames per second and allows sampling configuration. The best-practice recipe expands the video processor’s budget to roughly 224K video tokens for higher-rate hour-scale analysis; it is a configuration example rather than a hard universal clip-duration limit.
Thinking is enabled by default. The maker’s enable_thinking control can disable it for a direct answer. Qwen3.6 does not officially support the older Qwen3 soft prompt switches /think and /nothink. Separate the reasoning from the final artifact when presenting a result, and follow this checkpoint’s own interface rather than assume prompt markers or controls from another generation will apply.
Preserving historical thinking
By default, the template retains thinking blocks associated with the latest user message. Qwen3.6 has additionally been trained to use historical reasoning when preserve_thinking is enabled. Its explicit example sets that option to true. This can help an iterative agent task retain decision context instead of repeatedly re-deriving earlier choices; it also changes the amount of history the request carries. The availability of the option depends on the interface exposing the maker’s recipe.
Agentic usage
Qwen documents tool calling and recommends Qwen-Agent for configuring tools, including MCP-connected tools. Qwen Code is another maker-documented coding-agent workflow. Define the permitted tools, task goal and expected deliverable, then return actual tool observations to the model. The model can reason about a result and revise a plan; access to files or execution environments comes from the connected application, not the model name.
Ultra-long text
The documented context extension uses YaRN with the checkpoint’s interleaved positional settings. A factor of four targets the approximately one-million-token workflow; for 524,288-token workloads, Qwen suggests a factor of two. Static scaling can affect shorter-text performance, so the maker recommends changing it when the longer context is needed. Retain section and code identifiers in a large input and ask the result to show which evidence supports each conclusion.
Generation guidance and limits
For general thinking, the maker recommends temperature 1.0, top_p 0.95, top_k 20, min_p 0, presence penalty 1.5 and repetition penalty 1.0. For precise coding, temperature becomes 0.6 and presence penalty zero. For direct responses, it recommends temperature 0.7, top_p 0.8 and presence penalty 1.5, with the other values unchanged. Higher presence penalties can reduce repetition but may mix languages or slightly reduce quality.
Qwen recommends a 32,768-token generation allowance for most questions and 81,920 for unusually complex mathematical or programming tasks. These are recommendations, not separately established hard output ceilings. State the final format explicitly and check code, calculations and extracted visual details against the source. Reasoning continuity and long context help carry a task’s evidence, while a checkable final artifact makes the result useful.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits qwen/qwen3.6-35b-a3b | 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 $0.000566 / 1K tokens Output · Output tokens $0.00341 / 1K tokens Input · Cache read $0.0001 / 1K tokens Input · Cache write $0 / 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 $0.000566 / 1K tokens Output · Output tokens $0.00341 / 1K tokens Input · Cache read $0.0001 / 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": "qwen/qwen3.6-35b-a3b", "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='qwen/qwen3.6-35b-a3b',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='qwen/qwen3.6-35b-a3b',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 Ultra qwen/qwen3.6-35b-a3b
Use the method paths shown above.
qwen/qwen3.6-35b-a3bQwen3.6-35B-A3B is Qwen’s vision-language mixture-of-experts model with 35 billion total and 3 billion activated parameters. It focuses on frontend and repository-level coding, adds optional preservation of historical reasoning, and accepts text, images and video with textual output.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Qwen3.6 highlights: coding and reasoning continuity
Qwen3.6-35B-A3B is a vision-language model focused on useful coding and agent work. Qwen highlights stronger frontend workflows and repository-level reasoning, plus a new option to retain thinking from historical messages. It accepts text, images and video and returns text: an explanation, code proposal, structured analysis or another requested textual artifact. Preserve this exact 35B-A3B edition rather than treating it as the dense Qwen3.6-27B or its earlier Qwen3.5 counterpart.
For a frontend task, provide the relevant screenshot, functional requirements and existing code context. Ask for editable source and a description of the intended states and interactions. For a repository task, identify the affected feature, the input exposing a problem and expected behavior. Request a proposed patch with its assumptions and relevant checks.
Model overview
The mixture-of-experts model has 35 billion parameters in total and activates three billion. Its 40-layer hybrid layout repeats three Gated DeltaNet expert layers followed by one gated-attention expert layer. It has 256 routed experts and selects eight routed experts with one shared expert; multi-token prediction is part of training. The smaller activated count does not describe the full checkpoint size. A vision encoder supplies the visual part of the model.
The native context is 262,144 tokens, with conditional extension to 1,010,000. The maker recommends retaining a context of at least 128K when possible for complex thinking tasks, even if reducing it would ease memory pressure. Source text, represented visuals, historical messages, reasoning and the final answer share the configured sequence allowance.
Using inputs and response modes
The card shows text, image and video questions. For a diagram or page, preserve the labels and define the exact question. For a clip, keep timing and state the event of interest. The maker’s supported video example defaults to two frames per second and allows sampling configuration. The best-practice recipe expands the video processor’s budget to roughly 224K video tokens for higher-rate hour-scale analysis; it is a configuration example rather than a hard universal clip-duration limit.
Thinking is enabled by default. The maker’s enable_thinking control can disable it for a direct answer. Qwen3.6 does not officially support the older Qwen3 soft prompt switches /think and /nothink. Separate the reasoning from the final artifact when presenting a result, and follow this checkpoint’s own interface rather than assume prompt markers or controls from another generation will apply.
Preserving historical thinking
By default, the template retains thinking blocks associated with the latest user message. Qwen3.6 has additionally been trained to use historical reasoning when preserve_thinking is enabled. Its explicit example sets that option to true. This can help an iterative agent task retain decision context instead of repeatedly re-deriving earlier choices; it also changes the amount of history the request carries. The availability of the option depends on the interface exposing the maker’s recipe.
Agentic usage
Qwen documents tool calling and recommends Qwen-Agent for configuring tools, including MCP-connected tools. Qwen Code is another maker-documented coding-agent workflow. Define the permitted tools, task goal and expected deliverable, then return actual tool observations to the model. The model can reason about a result and revise a plan; access to files or execution environments comes from the connected application, not the model name.
Ultra-long text
The documented context extension uses YaRN with the checkpoint’s interleaved positional settings. A factor of four targets the approximately one-million-token workflow; for 524,288-token workloads, Qwen suggests a factor of two. Static scaling can affect shorter-text performance, so the maker recommends changing it when the longer context is needed. Retain section and code identifiers in a large input and ask the result to show which evidence supports each conclusion.
Generation guidance and limits
For general thinking, the maker recommends temperature 1.0, top_p 0.95, top_k 20, min_p 0, presence penalty 1.5 and repetition penalty 1.0. For precise coding, temperature becomes 0.6 and presence penalty zero. For direct responses, it recommends temperature 0.7, top_p 0.8 and presence penalty 1.5, with the other values unchanged. Higher presence penalties can reduce repetition but may mix languages or slightly reduce quality.
Qwen recommends a 32,768-token generation allowance for most questions and 81,920 for unusually complex mathematical or programming tasks. These are recommendations, not separately established hard output ceilings. State the final format explicitly and check code, calculations and extracted visual details against the source. Reasoning continuity and long context help carry a task’s evidence, while a checkable final artifact makes the result useful.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits qwen/qwen3.6-35b-a3b | 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 $0.000566 / 1K tokens Output · Output tokens $0.00341 / 1K tokens Input · Cache read $0.0001 / 1K tokens Input · Cache write $0 / 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 $0.000566 / 1K tokens Output · Output tokens $0.00341 / 1K tokens Input · Cache read $0.0001 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "qwen/qwen3.6-35b-a3b",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${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': 'qwen/qwen3.6-35b-a3b', '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': 'qwen/qwen3.6-35b-a3b', '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 qwen/qwen3.6-35b-a3b
Use the method paths shown above.
Technical example in English
Correct a button loading state\n\nPropose a small JavaScript change for a save form. Current handler sets button.disabled = true, awaits saveDraft(), then reenables the button. When saveDraft rejects, the button stays disabled. Preserve the save operation and show pseudocode for displaying an error and always restoring the button. Add success and failure acceptance checks.A focused async handler or pseudocode plus two checks covering button and error states.Verify the cleanup happens on rejection and success without converting a failed save into a success message.