Claude Sonnet 4.6 combines coding, understanding of large context, agentic planning, professional information work and design. Anthropic highlights studying code before making changes and the ability to work through multiple task steps in sequence. This is useful for finding a problem in a large project, analyzing a long document and preparing report content.
The model can analyze information from contracts, research papers and other material in one request. A task may include extracting facts, reasoning from them and forming a result that brings several parts together. Described uses also include financial analysis and creation of interface code with attention to composition and presentation.
Sonnet 4.6 understands text and images; its own output is text. Anthropic describes computer-use scenarios where the model navigates from a screenshot, chooses mouse and keyboard actions, analyzes a complex table or works through a sequence of web-form steps. Actual execution requires application tools that provide the screen and perform the actions.
The manufacturer's standard visual input formats are JPEG/PNG/GIF/WebP, using the first frame of an animation. For native analysis, the longer dimension is 1568 pixels, with 1568 visual tokens; larger images are downscaled before analysis. File and parameter support depends on the application's method.
Similarly, documents, spreadsheets and presentations can be analyzed or have their content prepared, but producing a finished file requires a suitable environment. Images help the model understand charts, diagrams and the visual structure of material; this does not turn it into an image generator.
This version supports adaptive reasoning and manual extended thinking. Manual mode is marked as deprecated; adaptive mode lets the model choose when additional analysis is useful, including between tool calls. The effort levels are low, medium, high and max, with high as the default. According to Anthropic, effort selection helps match the level of work and latency to the nature of the task.
The current specification provides 1 million tokens of context and up to 128 thousand response tokens in ordinary mode. For sustained work, the manufacturer also describes experimental compaction of old context into a summary. This feature requires application support and does not remove the model's context-window limits.
The reliable built-in knowledge cutoff is August 2025. Later information requires task material or available search. Connected tools can help find information, execute code and process discovered content; their availability and permitted actions are determined by the application.
Sonnet 4.6 was trained to resist hidden malicious instructions in external material, but complex visual environments and ambiguous tasks can still lead to errors. Explicit task requirements, verification of extracted facts and assessment of action results remain part of reliable work with the model.
", "aboutText": "### Code, documents and an action plan\n\nClaude Sonnet 4.6 combines coding, understanding of large context, agentic planning, professional information work and design. Anthropic highlights studying code before making changes and the ability to work through multiple task steps in sequence. This is useful for finding a problem in a large project, analyzing a long document and preparing report content.\n\nThe model can analyze information from contracts, research papers and other material in one request. A task may include extracting facts, reasoning from them and forming a result that brings several parts together. Described uses also include financial analysis and creation of interface code with attention to composition and presentation.\n\n### Actions in a visual environment\n\nSonnet 4.6 understands **text and images**; its own output is **text**. Anthropic describes computer-use scenarios where the model navigates from a screenshot, chooses mouse and keyboard actions, analyzes a complex table or works through a sequence of web-form steps. Actual execution requires application tools that provide the screen and perform the actions.\n\nThe manufacturer's standard visual input formats are JPEG/PNG/GIF/WebP, using the first frame of an animation. For native analysis, the longer dimension is 1568 pixels, with 1568 visual tokens; larger images are downscaled before analysis. File and parameter support depends on the application's method.\n\nSimilarly, documents, spreadsheets and presentations can be analyzed or have their content prepared, but producing a finished file requires a suitable environment. Images help the model understand charts, diagrams and the visual structure of material; this does not turn it into an image generator.\n\n### How analysis is controlled\n\nThis version supports adaptive reasoning and manual extended thinking. Manual mode is marked as deprecated; adaptive mode lets the model choose when additional analysis is useful, including between tool calls. The `effort` levels are `low`, `medium`, `high` and `max`, with `high` as the default. According to Anthropic, effort selection helps match the level of work and latency to the nature of the task.\n\nThe current specification provides **1 million tokens of context** and up to **128 thousand response tokens in ordinary mode**. For sustained work, the manufacturer also describes experimental compaction of old context into a summary. This feature requires application support and does not remove the model's context-window limits.\n\n### Points to consider when working\n\nThe reliable built-in knowledge cutoff is **August 2025**. Later information requires task material or available search. Connected tools can help find information, execute code and process discovered content; their availability and permitted actions are determined by the application.\n\nSonnet 4.6 was trained to resist hidden malicious instructions in external material, but complex visual environments and ambiguous tasks can still lead to errors. Explicit task requirements, verification of extracted facts and assessment of action results remain part of reliable work with the model.", "editorialOverlay": { "copy_fields": null, "copy_sha256": null, "copy_source": null, "html_sha256": "3d46fcf83ba9cbcebdac2c8b5fb189867795337af341b4ba828ff4fc5f0d00dc", "module_sha256": "ee9dc1c50b1e6a8fb7fa062030fcb4646b9aaaeb08be513b723a5a92a7290f06", "raw_model_unchanged": true, "schema": "ailoot-native-model-editorial-v1" }, "locale": "en", "methods": [ { "html": "claude-sonnet-4-6-thinkingClaude Sonnet 4.6 helps with development, understanding long documents, planning and tasks involving visual data. Supports adaptive reasoning and multi-step tool use.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Claude Sonnet 4.6 combines coding, understanding of large context, agentic planning, professional information work and design. Anthropic highlights studying code before making changes and the ability to work through multiple task steps in sequence. This is useful for finding a problem in a large project, analyzing a long document and preparing report content.
The model can analyze information from contracts, research papers and other material in one request. A task may include extracting facts, reasoning from them and forming a result that brings several parts together. Described uses also include financial analysis and creation of interface code with attention to composition and presentation.
Sonnet 4.6 understands text and images; its own output is text. Anthropic describes computer-use scenarios where the model navigates from a screenshot, chooses mouse and keyboard actions, analyzes a complex table or works through a sequence of web-form steps. Actual execution requires application tools that provide the screen and perform the actions.
The manufacturer's standard visual input formats are JPEG/PNG/GIF/WebP, using the first frame of an animation. For native analysis, the longer dimension is 1568 pixels, with 1568 visual tokens; larger images are downscaled before analysis. File and parameter support depends on the application's method.
Similarly, documents, spreadsheets and presentations can be analyzed or have their content prepared, but producing a finished file requires a suitable environment. Images help the model understand charts, diagrams and the visual structure of material; this does not turn it into an image generator.
This version supports adaptive reasoning and manual extended thinking. Manual mode is marked as deprecated; adaptive mode lets the model choose when additional analysis is useful, including between tool calls. The effort levels are low, medium, high and max, with high as the default. According to Anthropic, effort selection helps match the level of work and latency to the nature of the task.
The current specification provides 1 million tokens of context and up to 128 thousand response tokens in ordinary mode. For sustained work, the manufacturer also describes experimental compaction of old context into a summary. This feature requires application support and does not remove the model's context-window limits.
The reliable built-in knowledge cutoff is August 2025. Later information requires task material or available search. Connected tools can help find information, execute code and process discovered content; their availability and permitted actions are determined by the application.
Sonnet 4.6 was trained to resist hidden malicious instructions in external material, but complex visual environments and ambiguous tasks can still lead to errors. Explicit task requirements, verification of extracted facts and assessment of action results remain part of reliable work with the model.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits claude-sonnet-4-6-thinking | 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 $0.006 / 1K tokens Output · Output tokens $0.03 / 1K tokens Input · Cache read $0.0006 / 1K tokens Input · Cache write $0.0075 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model":"claude-sonnet-4-6-thinking","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="claude-sonnet-4-6-thinking", 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 claude-sonnet-4-6-thinking
/chat/completions