Ministral 3 3B Instruct 2512 is the smallest instruction-tuned edition of Mistral AI's Ministral 3 family. The maker positions it for compact instruction-based applications, including image captions, text classification, data extraction, translation and short content. The released checkpoint uses FP8 precision and combines text processing with a visual input component.
Model components and capabilities
The card lists a 3.4-billion-parameter language model and a 0.4-billion-parameter vision encoder. It documents a 256k context window, with a default configured length of 262,144 in the usage notes. The component counts belong to the 3B edition rather than the 8B or 14B family members. The model accepts text and images and generates text, including JSON-formatted records or proposed function calls.
The maker describes multilingual support across dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. It also highlights system-prompt adherence. Specify the required output language and the labels or terms that must remain unchanged. This keeps a translated explanation or extracted record aligned with the application's intended use.
Focused classification and extraction
For a classification task, provide the permitted labels, the rule distinguishing them and the message to classify. Ask for one selected label and a short explanation grounded in the supplied message. If the labels are ambiguous, define what information would distinguish them instead of relying on their names alone. This creates a compact result that can be used consistently in a larger workflow.
For extraction, list the fields and their types and state how to represent missing information. A request can ask for an item identifier, the condition explicitly described and the action the user is requesting. Require the source wording when exact wording matters, or a normalized value when the application supplies the normalization rule. JSON output organizes the result; the field design remains part of the task rather than a built-in universal schema.
The card also names image captioning and short content generation. For a caption, define the aspect of the image to describe and the audience. For a short explanation, provide the facts to preserve and the desired length. A support-style reply can use the original question and supplied guidance to produce a concise next step with the condition that makes it applicable.
Visual and multilingual preparation
Pair an image with a clear text question. For a document field, preserve the label, value and surrounding context needed to interpret it. For an object description, specify which visible feature matters. The maker recommends image proportions close to 1:1 and avoiding very thin or wide inputs; crop as needed without removing the evidence the question relies on.
For translation, provide the original text, the final language and any terminology list. Ask for the translated passage separately from an optional explanation. A multilingual classification task can keep application labels unchanged while explaining the selected label in the reader's language. These are defined outputs rather than a request to loosely discuss a message or image.
Tools and recommended settings
Native function calling enables an application to connect the conversation with a described operation. The model prepares the function arguments, the application executes the function and its result becomes information for the next answer. A lookup can supply a missing value, or a calculation tool can operate on values identified from an image. The text model itself does not execute the function simply by producing its name and arguments.
The maker recommends a system instruction defining the environment and use case, including tool guidance. Keep tool descriptions clear and the set limited to the minimum relevant operations. For a narrow assistant, an explicit extraction or lookup task can be easier to interpret than a prompt that includes many unrelated functions and no selection rule.
The card gives low-temperature guidance: below 0.1 for everyday and production-style use, and a starting temperature of 0.1 for most use cases in its separate sampling note. Higher values can be explored for creative content. These recommendations have different stated scopes; they are not a guarantee that one sampling setting makes every extracted value or classification correct.
Edition and context boundaries
Ministral 3 provides separate Base, Instruct and Reasoning releases at each model size. This entry identifies the Instruct 2512 checkpoint, rather than an automatically enabled reasoning mode or the pretrained base. The maker lists specialization and fine-tuning as possible applications, while the current instruction task still needs its input, rules and output clearly defined.
Keep relevant source material and conversation within the configured context and request the compact result the next application step needs. The usage notes allow reducing the configured context length for tasks that do not need the full window. A 512-token generation example and a large example generation allowance are not independently documented maximum output capacities. Image generation, audio output and external actions require other components beyond this text-result model.
", "aboutText": "Ministral 3 3B Instruct 2512 is the smallest instruction-tuned edition of Mistral AI's Ministral 3 family. The maker positions it for compact instruction-based applications, including image captions, text classification, data extraction, translation and short content. The released checkpoint uses FP8 precision and combines text processing with a visual input component.\n\nModel components and capabilities\n\nThe card lists a 3.4-billion-parameter language model and a 0.4-billion-parameter vision encoder. It documents a 256k context window, with a default configured length of 262,144 in the usage notes. The component counts belong to the 3B edition rather than the 8B or 14B family members. The model accepts text and images and generates text, including JSON-formatted records or proposed function calls.\n\nThe maker describes multilingual support across dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. It also highlights system-prompt adherence. Specify the required output language and the labels or terms that must remain unchanged. This keeps a translated explanation or extracted record aligned with the application's intended use.\n\nFocused classification and extraction\n\nFor a classification task, provide the permitted labels, the rule distinguishing them and the message to classify. Ask for one selected label and a short explanation grounded in the supplied message. If the labels are ambiguous, define what information would distinguish them instead of relying on their names alone. This creates a compact result that can be used consistently in a larger workflow.\n\nFor extraction, list the fields and their types and state how to represent missing information. A request can ask for an item identifier, the condition explicitly described and the action the user is requesting. Require the source wording when exact wording matters, or a normalized value when the application supplies the normalization rule. JSON output organizes the result; the field design remains part of the task rather than a built-in universal schema.\n\nThe card also names image captioning and short content generation. For a caption, define the aspect of the image to describe and the audience. For a short explanation, provide the facts to preserve and the desired length. A support-style reply can use the original question and supplied guidance to produce a concise next step with the condition that makes it applicable.\n\nVisual and multilingual preparation\n\nPair an image with a clear text question. For a document field, preserve the label, value and surrounding context needed to interpret it. For an object description, specify which visible feature matters. The maker recommends image proportions close to 1:1 and avoiding very thin or wide inputs; crop as needed without removing the evidence the question relies on.\n\nFor translation, provide the original text, the final language and any terminology list. Ask for the translated passage separately from an optional explanation. A multilingual classification task can keep application labels unchanged while explaining the selected label in the reader's language. These are defined outputs rather than a request to loosely discuss a message or image.\n\nTools and recommended settings\n\nNative function calling enables an application to connect the conversation with a described operation. The model prepares the function arguments, the application executes the function and its result becomes information for the next answer. A lookup can supply a missing value, or a calculation tool can operate on values identified from an image. The text model itself does not execute the function simply by producing its name and arguments.\n\nThe maker recommends a system instruction defining the environment and use case, including tool guidance. Keep tool descriptions clear and the set limited to the minimum relevant operations. For a narrow assistant, an explicit extraction or lookup task can be easier to interpret than a prompt that includes many unrelated functions and no selection rule.\n\nThe card gives low-temperature guidance: below 0.1 for everyday and production-style use, and a starting temperature of 0.1 for most use cases in its separate sampling note. Higher values can be explored for creative content. These recommendations have different stated scopes; they are not a guarantee that one sampling setting makes every extracted value or classification correct.\n\nEdition and context boundaries\n\nMinistral 3 provides separate Base, Instruct and Reasoning releases at each model size. This entry identifies the Instruct 2512 checkpoint, rather than an automatically enabled reasoning mode or the pretrained base. The maker lists specialization and fine-tuning as possible applications, while the current instruction task still needs its input, rules and output clearly defined.\n\nKeep relevant source material and conversation within the configured context and request the compact result the next application step needs. The usage notes allow reducing the configured context length for tasks that do not need the full window. A 512-token generation example and a large example generation allowance are not independently documented maximum output capacities. Image generation, audio output and external actions require other components beyond this text-result model.", "editorialOverlay": { "copy_fields": { "practical_html": "Technical example in English
Extract a return-request record\n\nExtract the item code, stated condition and requested action from this message: 'Hello, the handle of kettle KT8 is loose. It arrived today. Please arrange an exchange, not a refund.' Return those three fields and quote the wording supporting the requested action. Use 'not supplied' for any field absent from the message.A compact three-field extraction record plus the supporting request quote.Check that exchange is retained as the request and is not rewritten as a completed action.mistralai/ministral-3b-2512Ministral 3 3B Instruct 2512 is a compact multilingual text-and-image model for classification, extraction, image captions and short content. Its 3.4B language component and 0.4B vision encoder support system instructions, JSON, function calls and a 256k maker context.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Ministral 3 3B Instruct 2512 is the smallest instruction-tuned edition of Mistral AI's Ministral 3 family. The maker positions it for compact instruction-based applications, including image captions, text classification, data extraction, translation and short content. The released checkpoint uses FP8 precision and combines text processing with a visual input component.
Model components and capabilities
The card lists a 3.4-billion-parameter language model and a 0.4-billion-parameter vision encoder. It documents a 256k context window, with a default configured length of 262,144 in the usage notes. The component counts belong to the 3B edition rather than the 8B or 14B family members. The model accepts text and images and generates text, including JSON-formatted records or proposed function calls.
The maker describes multilingual support across dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. It also highlights system-prompt adherence. Specify the required output language and the labels or terms that must remain unchanged. This keeps a translated explanation or extracted record aligned with the application's intended use.
Focused classification and extraction
For a classification task, provide the permitted labels, the rule distinguishing them and the message to classify. Ask for one selected label and a short explanation grounded in the supplied message. If the labels are ambiguous, define what information would distinguish them instead of relying on their names alone. This creates a compact result that can be used consistently in a larger workflow.
For extraction, list the fields and their types and state how to represent missing information. A request can ask for an item identifier, the condition explicitly described and the action the user is requesting. Require the source wording when exact wording matters, or a normalized value when the application supplies the normalization rule. JSON output organizes the result; the field design remains part of the task rather than a built-in universal schema.
The card also names image captioning and short content generation. For a caption, define the aspect of the image to describe and the audience. For a short explanation, provide the facts to preserve and the desired length. A support-style reply can use the original question and supplied guidance to produce a concise next step with the condition that makes it applicable.
Visual and multilingual preparation
Pair an image with a clear text question. For a document field, preserve the label, value and surrounding context needed to interpret it. For an object description, specify which visible feature matters. The maker recommends image proportions close to 1:1 and avoiding very thin or wide inputs; crop as needed without removing the evidence the question relies on.
For translation, provide the original text, the final language and any terminology list. Ask for the translated passage separately from an optional explanation. A multilingual classification task can keep application labels unchanged while explaining the selected label in the reader's language. These are defined outputs rather than a request to loosely discuss a message or image.
Tools and recommended settings
Native function calling enables an application to connect the conversation with a described operation. The model prepares the function arguments, the application executes the function and its result becomes information for the next answer. A lookup can supply a missing value, or a calculation tool can operate on values identified from an image. The text model itself does not execute the function simply by producing its name and arguments.
The maker recommends a system instruction defining the environment and use case, including tool guidance. Keep tool descriptions clear and the set limited to the minimum relevant operations. For a narrow assistant, an explicit extraction or lookup task can be easier to interpret than a prompt that includes many unrelated functions and no selection rule.
The card gives low-temperature guidance: below 0.1 for everyday and production-style use, and a starting temperature of 0.1 for most use cases in its separate sampling note. Higher values can be explored for creative content. These recommendations have different stated scopes; they are not a guarantee that one sampling setting makes every extracted value or classification correct.
Edition and context boundaries
Ministral 3 provides separate Base, Instruct and Reasoning releases at each model size. This entry identifies the Instruct 2512 checkpoint, rather than an automatically enabled reasoning mode or the pretrained base. The maker lists specialization and fine-tuning as possible applications, while the current instruction task still needs its input, rules and output clearly defined.
Keep relevant source material and conversation within the configured context and request the compact result the next application step needs. The usage notes allow reducing the configured context length for tasks that do not need the full window. A 512-token generation example and a large example generation allowance are not independently documented maximum output capacities. Image generation, audio output and external actions require other components beyond this text-result model.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits mistralai/ministral-3b-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 $0.00066 / 1K tokens Output · Output tokens $0.00066 / 1K tokens Input · Cache read $0.00002 / 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 $0.00066 / 1K tokens Output · Output tokens $0.00066 / 1K tokens Input · Cache read $0.00002 / 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/ministral-3b-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/ministral-3b-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/ministral-3b-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/ministral-3b-2512
Use the method paths shown above.
mistralai/ministral-3b-2512Ministral 3 3B Instruct 2512 is a compact multilingual text-and-image model for classification, extraction, image captions and short content. Its 3.4B language component and 0.4B vision encoder support system instructions, JSON, function calls and a 256k maker context.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Ministral 3 3B Instruct 2512 is the smallest instruction-tuned edition of Mistral AI's Ministral 3 family. The maker positions it for compact instruction-based applications, including image captions, text classification, data extraction, translation and short content. The released checkpoint uses FP8 precision and combines text processing with a visual input component.
Model components and capabilities
The card lists a 3.4-billion-parameter language model and a 0.4-billion-parameter vision encoder. It documents a 256k context window, with a default configured length of 262,144 in the usage notes. The component counts belong to the 3B edition rather than the 8B or 14B family members. The model accepts text and images and generates text, including JSON-formatted records or proposed function calls.
The maker describes multilingual support across dozens of languages, including English, French, Spanish, German, Italian, Portuguese, Dutch, Chinese, Japanese, Korean and Arabic. It also highlights system-prompt adherence. Specify the required output language and the labels or terms that must remain unchanged. This keeps a translated explanation or extracted record aligned with the application's intended use.
Focused classification and extraction
For a classification task, provide the permitted labels, the rule distinguishing them and the message to classify. Ask for one selected label and a short explanation grounded in the supplied message. If the labels are ambiguous, define what information would distinguish them instead of relying on their names alone. This creates a compact result that can be used consistently in a larger workflow.
For extraction, list the fields and their types and state how to represent missing information. A request can ask for an item identifier, the condition explicitly described and the action the user is requesting. Require the source wording when exact wording matters, or a normalized value when the application supplies the normalization rule. JSON output organizes the result; the field design remains part of the task rather than a built-in universal schema.
The card also names image captioning and short content generation. For a caption, define the aspect of the image to describe and the audience. For a short explanation, provide the facts to preserve and the desired length. A support-style reply can use the original question and supplied guidance to produce a concise next step with the condition that makes it applicable.
Visual and multilingual preparation
Pair an image with a clear text question. For a document field, preserve the label, value and surrounding context needed to interpret it. For an object description, specify which visible feature matters. The maker recommends image proportions close to 1:1 and avoiding very thin or wide inputs; crop as needed without removing the evidence the question relies on.
For translation, provide the original text, the final language and any terminology list. Ask for the translated passage separately from an optional explanation. A multilingual classification task can keep application labels unchanged while explaining the selected label in the reader's language. These are defined outputs rather than a request to loosely discuss a message or image.
Tools and recommended settings
Native function calling enables an application to connect the conversation with a described operation. The model prepares the function arguments, the application executes the function and its result becomes information for the next answer. A lookup can supply a missing value, or a calculation tool can operate on values identified from an image. The text model itself does not execute the function simply by producing its name and arguments.
The maker recommends a system instruction defining the environment and use case, including tool guidance. Keep tool descriptions clear and the set limited to the minimum relevant operations. For a narrow assistant, an explicit extraction or lookup task can be easier to interpret than a prompt that includes many unrelated functions and no selection rule.
The card gives low-temperature guidance: below 0.1 for everyday and production-style use, and a starting temperature of 0.1 for most use cases in its separate sampling note. Higher values can be explored for creative content. These recommendations have different stated scopes; they are not a guarantee that one sampling setting makes every extracted value or classification correct.
Edition and context boundaries
Ministral 3 provides separate Base, Instruct and Reasoning releases at each model size. This entry identifies the Instruct 2512 checkpoint, rather than an automatically enabled reasoning mode or the pretrained base. The maker lists specialization and fine-tuning as possible applications, while the current instruction task still needs its input, rules and output clearly defined.
Keep relevant source material and conversation within the configured context and request the compact result the next application step needs. The usage notes allow reducing the configured context length for tasks that do not need the full window. A 512-token generation example and a large example generation allowance are not independently documented maximum output capacities. Image generation, audio output and external actions require other components beyond this text-result model.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits mistralai/ministral-3b-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 $0.00066 / 1K tokens Output · Output tokens $0.00066 / 1K tokens Input · Cache read $0.00002 / 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 $0.00066 / 1K tokens Output · Output tokens $0.00066 / 1K tokens Input · Cache read $0.00002 / 1K tokens Input · Cache write $0 / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "mistralai/ministral-3b-2512",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${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/ministral-3b-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/ministral-3b-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/ministral-3b-2512
Use the method paths shown above.
Technical example in English
Extract a return-request record\n\nExtract the item code, stated condition and requested action from this message: 'Hello, the handle of kettle KT8 is loose. It arrived today. Please arrange an exchange, not a refund.' Return those three fields and quote the wording supporting the requested action. Use 'not supplied' for any field absent from the message.A compact three-field extraction record plus the supporting request quote.Check that exchange is retained as the request and is not rewritten as a completed action.