Mistral Large 3 2512 is the instruction-tuned Mistral AI edition for general conversation, visual understanding and enterprise work. Its sparse mixture-of-experts architecture has 675 billion total parameters, with 41 billion active, and a 256k context window. Text and images can contribute to the same task, while the answer is text, code or a structured result.
Capabilities and input preparation
The model combines a language component with a vision encoder. Give an image a clear role: a page to read, a diagram to interpret or a screen to explain. For a crowded screen, provide a crop containing the relevant controls and retain the surrounding labels needed to understand them. Mistral recommends image shapes close to square; break an exceptionally narrow or wide source into useful regions instead of shrinking important details into an unreadable strip.
The maker lists multilingual conversation across dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. State the output language, intended reader and terminology to preserve. For a support article, ask for a localized explanation with product names unchanged and a short list of uncertain translations to resolve before publication.
Long documents and enterprise knowledge
Use the context to bring related material together: policy text, a specification, existing decisions and the question that needs an answer. Label each source and its revision. A useful request is: “Compare these three onboarding policies. Return a table of shared requirements, conflicts and unanswered questions, with the source section for each finding.” The resulting artifact is a comparison grounded in the supplied passages rather than a generic summary of the topic.
For a document question, identify the desired level of detail and request the supporting passage next to the conclusion. Mistral's Document QnA workflow first extracts text and document structure with OCR, then uses a language model to interpret that material. Preserve page numbers and headings when preparing extracted text so the answer can point back to the original. A document URL, extracted text and image evidence have different preparation requirements; choose the input form supported by the application you are using.
Coding and everyday assistance
Mistral positions Large 3 for coding assistance as well as daily conversation. Supply the relevant functions, expected behavior and observed failure. For example: “This parser drops the last field when the input ends without a newline. Explain the failing branch, propose the smallest change and return the corrected function with three representative input/output examples.” The useful result is a proposed change tied to the actual code and a set of cases the developer can run.
For a business draft, provide the facts and desired artifact together. A request to turn meeting notes into a decision memo should name the decision, alternatives, unresolved assumptions and next action. Ask the model to keep source facts separate from recommendations and to mark missing information within the appropriate section. This makes the output easier to revise than an undifferentiated narrative.
Tools and structured results
Native function calling lets the model choose a supplied function and prepare its arguments. The application runs that function and returns its result; the model then uses the returned information in the answer. Keep tools specific and expose only those needed for the task. A stock-check assistant can receive a product reference, request a stock lookup and explain the returned quantity and warehouse. Validate the selected function and arguments before executing the request, and retain the tool response alongside the final answer.
For extraction, specify the fields, types and treatment of missing values. Mistral documents JSON mode and custom structured output; its JSON mode requires an explicit instruction to return JSON, while a custom schema supplies stronger format control. An invoice summary can ask for supplier, invoice number, due date, currency and line items, with missing values represented consistently. Inspect the extracted values against the document, even when the response matches the requested structure.
Settings and edition differences
Define the role, task and tool rules in the system instructions. Mistral recommends a temperature below 0.1 for production and everyday use of this checkpoint, with higher values explored for creative tasks. Separate a request for varied ideas from a request for a reproducible extraction: give the former room for alternatives and the latter explicit fields and source constraints.
This instruction edition is separate from the Large 3 base checkpoint and from dedicated reasoning models. The maker identifies strict reasoning and specialized visual work as areas where a model designed specifically for those tasks can be preferable. For a difficult calculation, supply the assumptions and units, ask for intermediate checks and use an available calculation tool to verify the result. For dense visual evidence, prepare readable regions and compare the answer with the source before using it in a report.
", "aboutText": "Mistral Large 3 2512 is the instruction-tuned Mistral AI edition for general conversation, visual understanding and enterprise work. Its sparse mixture-of-experts architecture has 675 billion total parameters, with 41 billion active, and a 256k context window. Text and images can contribute to the same task, while the answer is text, code or a structured result.\n\nCapabilities and input preparation\n\nThe model combines a language component with a vision encoder. Give an image a clear role: a page to read, a diagram to interpret or a screen to explain. For a crowded screen, provide a crop containing the relevant controls and retain the surrounding labels needed to understand them. Mistral recommends image shapes close to square; break an exceptionally narrow or wide source into useful regions instead of shrinking important details into an unreadable strip.\n\nThe maker lists multilingual conversation across dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. State the output language, intended reader and terminology to preserve. For a support article, ask for a localized explanation with product names unchanged and a short list of uncertain translations to resolve before publication.\n\nLong documents and enterprise knowledge\n\nUse the context to bring related material together: policy text, a specification, existing decisions and the question that needs an answer. Label each source and its revision. A useful request is: “Compare these three onboarding policies. Return a table of shared requirements, conflicts and unanswered questions, with the source section for each finding.” The resulting artifact is a comparison grounded in the supplied passages rather than a generic summary of the topic.\n\nFor a document question, identify the desired level of detail and request the supporting passage next to the conclusion. Mistral's Document QnA workflow first extracts text and document structure with OCR, then uses a language model to interpret that material. Preserve page numbers and headings when preparing extracted text so the answer can point back to the original. A document URL, extracted text and image evidence have different preparation requirements; choose the input form supported by the application you are using.\n\nCoding and everyday assistance\n\nMistral positions Large 3 for coding assistance as well as daily conversation. Supply the relevant functions, expected behavior and observed failure. For example: “This parser drops the last field when the input ends without a newline. Explain the failing branch, propose the smallest change and return the corrected function with three representative input/output examples.” The useful result is a proposed change tied to the actual code and a set of cases the developer can run.\n\nFor a business draft, provide the facts and desired artifact together. A request to turn meeting notes into a decision memo should name the decision, alternatives, unresolved assumptions and next action. Ask the model to keep source facts separate from recommendations and to mark missing information within the appropriate section. This makes the output easier to revise than an undifferentiated narrative.\n\nTools and structured results\n\nNative function calling lets the model choose a supplied function and prepare its arguments. The application runs that function and returns its result; the model then uses the returned information in the answer. Keep tools specific and expose only those needed for the task. A stock-check assistant can receive a product reference, request a stock lookup and explain the returned quantity and warehouse. Validate the selected function and arguments before executing the request, and retain the tool response alongside the final answer.\n\nFor extraction, specify the fields, types and treatment of missing values. Mistral documents JSON mode and custom structured output; its JSON mode requires an explicit instruction to return JSON, while a custom schema supplies stronger format control. An invoice summary can ask for supplier, invoice number, due date, currency and line items, with missing values represented consistently. Inspect the extracted values against the document, even when the response matches the requested structure.\n\nSettings and edition differences\n\nDefine the role, task and tool rules in the system instructions. Mistral recommends a temperature below 0.1 for production and everyday use of this checkpoint, with higher values explored for creative tasks. Separate a request for varied ideas from a request for a reproducible extraction: give the former room for alternatives and the latter explicit fields and source constraints.\n\nThis instruction edition is separate from the Large 3 base checkpoint and from dedicated reasoning models. The maker identifies strict reasoning and specialized visual work as areas where a model designed specifically for those tasks can be preferable. For a difficult calculation, supply the assumptions and units, ask for intermediate checks and use an available calculation tool to verify the result. For dense visual evidence, prepare readable regions and compare the answer with the source before using it in a report.", "editorialOverlay": { "copy_fields": { "practical_html": "Technical example in English
Compare onboarding policies\n\nCompare onboarding policies. A section 1 requires manager approval and identity verification before access. B section 3 requires manager approval before access and identity verification within seven days afterward. List the shared requirement, the conflict and a question for the policy owner. Keep the source section beside each finding.A source-linked comparison of shared and conflicting requirements plus one policy decision question.Check that both policies’ timing for identity verification remains explicit and that neither is silently treated as authoritative.mistralai/mistral-large-2512Mistral Large 3 2512 is a multilingual text-and-image model for long documents, enterprise knowledge, coding and tool-assisted work. Its 256k context, native function calling and structured output help turn supplied evidence into answers, records and proposed actions.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Mistral Large 3 2512 is the instruction-tuned Mistral AI edition for general conversation, visual understanding and enterprise work. Its sparse mixture-of-experts architecture has 675 billion total parameters, with 41 billion active, and a 256k context window. Text and images can contribute to the same task, while the answer is text, code or a structured result.
Capabilities and input preparation
The model combines a language component with a vision encoder. Give an image a clear role: a page to read, a diagram to interpret or a screen to explain. For a crowded screen, provide a crop containing the relevant controls and retain the surrounding labels needed to understand them. Mistral recommends image shapes close to square; break an exceptionally narrow or wide source into useful regions instead of shrinking important details into an unreadable strip.
The maker lists multilingual conversation across dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. State the output language, intended reader and terminology to preserve. For a support article, ask for a localized explanation with product names unchanged and a short list of uncertain translations to resolve before publication.
Long documents and enterprise knowledge
Use the context to bring related material together: policy text, a specification, existing decisions and the question that needs an answer. Label each source and its revision. A useful request is: “Compare these three onboarding policies. Return a table of shared requirements, conflicts and unanswered questions, with the source section for each finding.” The resulting artifact is a comparison grounded in the supplied passages rather than a generic summary of the topic.
For a document question, identify the desired level of detail and request the supporting passage next to the conclusion. Mistral's Document QnA workflow first extracts text and document structure with OCR, then uses a language model to interpret that material. Preserve page numbers and headings when preparing extracted text so the answer can point back to the original. A document URL, extracted text and image evidence have different preparation requirements; choose the input form supported by the application you are using.
Coding and everyday assistance
Mistral positions Large 3 for coding assistance as well as daily conversation. Supply the relevant functions, expected behavior and observed failure. For example: “This parser drops the last field when the input ends without a newline. Explain the failing branch, propose the smallest change and return the corrected function with three representative input/output examples.” The useful result is a proposed change tied to the actual code and a set of cases the developer can run.
For a business draft, provide the facts and desired artifact together. A request to turn meeting notes into a decision memo should name the decision, alternatives, unresolved assumptions and next action. Ask the model to keep source facts separate from recommendations and to mark missing information within the appropriate section. This makes the output easier to revise than an undifferentiated narrative.
Tools and structured results
Native function calling lets the model choose a supplied function and prepare its arguments. The application runs that function and returns its result; the model then uses the returned information in the answer. Keep tools specific and expose only those needed for the task. A stock-check assistant can receive a product reference, request a stock lookup and explain the returned quantity and warehouse. Validate the selected function and arguments before executing the request, and retain the tool response alongside the final answer.
For extraction, specify the fields, types and treatment of missing values. Mistral documents JSON mode and custom structured output; its JSON mode requires an explicit instruction to return JSON, while a custom schema supplies stronger format control. An invoice summary can ask for supplier, invoice number, due date, currency and line items, with missing values represented consistently. Inspect the extracted values against the document, even when the response matches the requested structure.
Settings and edition differences
Define the role, task and tool rules in the system instructions. Mistral recommends a temperature below 0.1 for production and everyday use of this checkpoint, with higher values explored for creative tasks. Separate a request for varied ideas from a request for a reproducible extraction: give the former room for alternatives and the latter explicit fields and source constraints.
This instruction edition is separate from the Large 3 base checkpoint and from dedicated reasoning models. The maker identifies strict reasoning and specialized visual work as areas where a model designed specifically for those tasks can be preferable. For a difficult calculation, supply the assumptions and units, ask for intermediate checks and use an available calculation tool to verify the result. For dense visual evidence, prepare readable regions and compare the answer with the source before using it in a report.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits mistralai/mistral-large-2512 | 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 $0.0022 / 1K tokens Output · Output tokens $0.0066 / 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 $0.0022 / 1K tokens Output · Output tokens $0.0066 / 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": "mistralai/mistral-large-2512", "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='mistralai/mistral-large-2512',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='mistralai/mistral-large-2512',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 mistralai/mistral-large-2512
Use the method paths shown above.
mistralai/mistral-large-2512Mistral Large 3 2512 is a multilingual text-and-image model for long documents, enterprise knowledge, coding and tool-assisted work. Its 256k context, native function calling and structured output help turn supplied evidence into answers, records and proposed actions.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Mistral Large 3 2512 is the instruction-tuned Mistral AI edition for general conversation, visual understanding and enterprise work. Its sparse mixture-of-experts architecture has 675 billion total parameters, with 41 billion active, and a 256k context window. Text and images can contribute to the same task, while the answer is text, code or a structured result.
Capabilities and input preparation
The model combines a language component with a vision encoder. Give an image a clear role: a page to read, a diagram to interpret or a screen to explain. For a crowded screen, provide a crop containing the relevant controls and retain the surrounding labels needed to understand them. Mistral recommends image shapes close to square; break an exceptionally narrow or wide source into useful regions instead of shrinking important details into an unreadable strip.
The maker lists multilingual conversation across dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. State the output language, intended reader and terminology to preserve. For a support article, ask for a localized explanation with product names unchanged and a short list of uncertain translations to resolve before publication.
Long documents and enterprise knowledge
Use the context to bring related material together: policy text, a specification, existing decisions and the question that needs an answer. Label each source and its revision. A useful request is: “Compare these three onboarding policies. Return a table of shared requirements, conflicts and unanswered questions, with the source section for each finding.” The resulting artifact is a comparison grounded in the supplied passages rather than a generic summary of the topic.
For a document question, identify the desired level of detail and request the supporting passage next to the conclusion. Mistral's Document QnA workflow first extracts text and document structure with OCR, then uses a language model to interpret that material. Preserve page numbers and headings when preparing extracted text so the answer can point back to the original. A document URL, extracted text and image evidence have different preparation requirements; choose the input form supported by the application you are using.
Coding and everyday assistance
Mistral positions Large 3 for coding assistance as well as daily conversation. Supply the relevant functions, expected behavior and observed failure. For example: “This parser drops the last field when the input ends without a newline. Explain the failing branch, propose the smallest change and return the corrected function with three representative input/output examples.” The useful result is a proposed change tied to the actual code and a set of cases the developer can run.
For a business draft, provide the facts and desired artifact together. A request to turn meeting notes into a decision memo should name the decision, alternatives, unresolved assumptions and next action. Ask the model to keep source facts separate from recommendations and to mark missing information within the appropriate section. This makes the output easier to revise than an undifferentiated narrative.
Tools and structured results
Native function calling lets the model choose a supplied function and prepare its arguments. The application runs that function and returns its result; the model then uses the returned information in the answer. Keep tools specific and expose only those needed for the task. A stock-check assistant can receive a product reference, request a stock lookup and explain the returned quantity and warehouse. Validate the selected function and arguments before executing the request, and retain the tool response alongside the final answer.
For extraction, specify the fields, types and treatment of missing values. Mistral documents JSON mode and custom structured output; its JSON mode requires an explicit instruction to return JSON, while a custom schema supplies stronger format control. An invoice summary can ask for supplier, invoice number, due date, currency and line items, with missing values represented consistently. Inspect the extracted values against the document, even when the response matches the requested structure.
Settings and edition differences
Define the role, task and tool rules in the system instructions. Mistral recommends a temperature below 0.1 for production and everyday use of this checkpoint, with higher values explored for creative tasks. Separate a request for varied ideas from a request for a reproducible extraction: give the former room for alternatives and the latter explicit fields and source constraints.
This instruction edition is separate from the Large 3 base checkpoint and from dedicated reasoning models. The maker identifies strict reasoning and specialized visual work as areas where a model designed specifically for those tasks can be preferable. For a difficult calculation, supply the assumptions and units, ask for intermediate checks and use an available calculation tool to verify the result. For dense visual evidence, prepare readable regions and compare the answer with the source before using it in a report.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits mistralai/mistral-large-2512 | 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 $0.0022 / 1K tokens Output · Output tokens $0.0066 / 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 $0.0022 / 1K tokens Output · Output tokens $0.0066 / 1K tokens Input · Cache read $0.0001 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "mistralai/mistral-large-2512",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${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': 'mistralai/mistral-large-2512', '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': 'mistralai/mistral-large-2512', '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 mistralai/mistral-large-2512
Use the method paths shown above.
Technical example in English
Compare onboarding policies\n\nCompare onboarding policies. A section 1 requires manager approval and identity verification before access. B section 3 requires manager approval before access and identity verification within seven days afterward. List the shared requirement, the conflict and a question for the policy owner. Keep the source section beside each finding.A source-linked comparison of shared and conflicting requirements plus one policy decision question.Check that both policies’ timing for identity verification remains explicit and that neither is silently treated as authoritative.