Mistral Small 3.2 24B Instruct 2506 is a targeted update of Mistral Small 3.1 24B Instruct 2503. The maker identifies three specific changes: following precise instructions more closely, reducing repetitive or runaway answers, and making the function-calling template more robust. The model remains an instruction-tuned text-and-image model that generates text rather than a new audio or image-generation edition.
What changes and what carries forward
The 3.2 card explicitly states that its key features are the same as the named Small 3.1 release. Those features include vision, multilingual work, system-prompt support, native function calling, JSON output and a 128k context window. The referenced 3.1 card describes a 24-billion-parameter model and a Tekken tokenizer with a vocabulary labeled 131k. This documented link supports carrying those features into 3.2 while preserving the 2506 update as its own edition.
The instruction-following focus matters when the result has several exact requirements. Give the source material, list the rules the output must satisfy and state the artifact to return. A document-extraction request might require specified fields, original identifiers, a defined missing-value representation and no prose outside the record. A writing task might require a particular audience and section order while keeping supplied conditions unchanged.
The repetition change addresses generation behavior rather than adding a new reasoning mode. For a long answer, define the required sections and the purpose of each so that the request has a clear stopping point. The function-template change concerns how tool-enabled conversation is represented; use the matching edition's conversation and tokenizer settings rather than treating an older formatting example as interchangeable.
Visual and multilingual tasks
The inherited vision feature allows a text question to be paired with an image. Supply a readable page, chart or diagram and identify the detail needed by the task. For a chart comparison, retain the axes, units and legend and request the observation supporting each conclusion. For a document field, keep the label and surrounding context so the returned value can be interpreted correctly.
The documented language coverage includes Russian, English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, Farsi, Hindi, Bengali and other languages listed by the maker. Specify the final language and any terminology that should remain unchanged. A bilingual document task can preserve identifiers and units while adapting the surrounding explanation to the reader.
For a practical visual extraction, ask for the document identifier, field label, extracted text and the part of the image supporting it. An explanation can follow separately if the reader needs interpretation. This keeps transcription, structured fields and the model's conclusion from becoming one indistinguishable result.
System instructions and tools
The 3.2 card recommends a system prompt tailored to the use case and a relatively low temperature, such as 0.15. Set the assistant's role, response rules and treatment of source material explicitly. Place the actual document or code with the user task, preserving the boundary between instructions about the answer and instructions quoted inside the source.
Function calling needs the application to supply tool descriptions and required arguments. The model prepares a call, the application executes the function and returns its result, and the next answer uses that information. The released examples include visual reasoning, tool calls and text-only instructions. A lookup task can therefore ask the model to select a described operation while leaving the actual lookup and its result to the surrounding application.
For a coding request, provide the relevant function, the input exposing the issue and the behavior that should result. Ask for a proposed correction with the original interface preserved, then examples that distinguish the intended behavior. For a mathematical request, define variables and assumptions and ask for the important justified steps and a separate final expression. These are applications of the inherited programming and reasoning capabilities; execution of proposed code remains a separate step.
Context and edition boundaries
Use the 128k window for the source passages, images, conversation and expected answer together. A large-document comparison can retain related sections while returning a short matrix of findings. A large configured image allowance or generation value appearing in a serving example is not an independently stated maximum capacity of the model.
Small 3.2 is an instruction update with documented visual input, not the text-only Small 3 2501 or the hybrid Small 4 reasoning release. The maker's card does not provide a separate hard output-token ceiling for this entry. Preserve its exact 2506 identity and formatting, and define the structured record, explanation or code proposal needed by the next step.
", "aboutText": "Mistral Small 3.2 24B Instruct 2506 is a targeted update of Mistral Small 3.1 24B Instruct 2503. The maker identifies three specific changes: following precise instructions more closely, reducing repetitive or runaway answers, and making the function-calling template more robust. The model remains an instruction-tuned text-and-image model that generates text rather than a new audio or image-generation edition.\n\nWhat changes and what carries forward\n\nThe 3.2 card explicitly states that its key features are the same as the named Small 3.1 release. Those features include vision, multilingual work, system-prompt support, native function calling, JSON output and a 128k context window. The referenced 3.1 card describes a 24-billion-parameter model and a Tekken tokenizer with a vocabulary labeled 131k. This documented link supports carrying those features into 3.2 while preserving the 2506 update as its own edition.\n\nThe instruction-following focus matters when the result has several exact requirements. Give the source material, list the rules the output must satisfy and state the artifact to return. A document-extraction request might require specified fields, original identifiers, a defined missing-value representation and no prose outside the record. A writing task might require a particular audience and section order while keeping supplied conditions unchanged.\n\nThe repetition change addresses generation behavior rather than adding a new reasoning mode. For a long answer, define the required sections and the purpose of each so that the request has a clear stopping point. The function-template change concerns how tool-enabled conversation is represented; use the matching edition's conversation and tokenizer settings rather than treating an older formatting example as interchangeable.\n\nVisual and multilingual tasks\n\nThe inherited vision feature allows a text question to be paired with an image. Supply a readable page, chart or diagram and identify the detail needed by the task. For a chart comparison, retain the axes, units and legend and request the observation supporting each conclusion. For a document field, keep the label and surrounding context so the returned value can be interpreted correctly.\n\nThe documented language coverage includes Russian, English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, Farsi, Hindi, Bengali and other languages listed by the maker. Specify the final language and any terminology that should remain unchanged. A bilingual document task can preserve identifiers and units while adapting the surrounding explanation to the reader.\n\nFor a practical visual extraction, ask for the document identifier, field label, extracted text and the part of the image supporting it. An explanation can follow separately if the reader needs interpretation. This keeps transcription, structured fields and the model's conclusion from becoming one indistinguishable result.\n\nSystem instructions and tools\n\nThe 3.2 card recommends a system prompt tailored to the use case and a relatively low temperature, such as 0.15. Set the assistant's role, response rules and treatment of source material explicitly. Place the actual document or code with the user task, preserving the boundary between instructions about the answer and instructions quoted inside the source.\n\nFunction calling needs the application to supply tool descriptions and required arguments. The model prepares a call, the application executes the function and returns its result, and the next answer uses that information. The released examples include visual reasoning, tool calls and text-only instructions. A lookup task can therefore ask the model to select a described operation while leaving the actual lookup and its result to the surrounding application.\n\nFor a coding request, provide the relevant function, the input exposing the issue and the behavior that should result. Ask for a proposed correction with the original interface preserved, then examples that distinguish the intended behavior. For a mathematical request, define variables and assumptions and ask for the important justified steps and a separate final expression. These are applications of the inherited programming and reasoning capabilities; execution of proposed code remains a separate step.\n\nContext and edition boundaries\n\nUse the 128k window for the source passages, images, conversation and expected answer together. A large-document comparison can retain related sections while returning a short matrix of findings. A large configured image allowance or generation value appearing in a serving example is not an independently stated maximum capacity of the model.\n\nSmall 3.2 is an instruction update with documented visual input, not the text-only Small 3 2501 or the hybrid Small 4 reasoning release. The maker's card does not provide a separate hard output-token ceiling for this entry. Preserve its exact 2506 identity and formatting, and define the structured record, explanation or code proposal needed by the next step.", "editorialOverlay": { "copy_fields": { "practical_html": "Technical example in English
Extract exactly the requested fields\n\nRead this note: 'Service case C-52. Customer: Dana Lee. Device: Lamp Mini. Reported problem: switches off after ten minutes.' Return exactly four labelled lines, in this order: Case ID, Device, Problem, Purchase date. Use unknown for missing information. Omit the customer's name and all introductory or closing prose.Exactly four labelled lines containing source-derived values or unknown, in the requested order.Purchase date is unknown, Dana Lee is omitted, and there are no extra lines.mistralai/mistral-small-3.2-24b-instructMistral Small 3.2 24B Instruct is a text-and-image instruction model with a 128k maker context. Its 2506 update specifically improves precise instruction following, reduces repetition and strengthens the function-calling template while retaining Small 3.1's core features.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Mistral Small 3.2 24B Instruct 2506 is a targeted update of Mistral Small 3.1 24B Instruct 2503. The maker identifies three specific changes: following precise instructions more closely, reducing repetitive or runaway answers, and making the function-calling template more robust. The model remains an instruction-tuned text-and-image model that generates text rather than a new audio or image-generation edition.
What changes and what carries forward
The 3.2 card explicitly states that its key features are the same as the named Small 3.1 release. Those features include vision, multilingual work, system-prompt support, native function calling, JSON output and a 128k context window. The referenced 3.1 card describes a 24-billion-parameter model and a Tekken tokenizer with a vocabulary labeled 131k. This documented link supports carrying those features into 3.2 while preserving the 2506 update as its own edition.
The instruction-following focus matters when the result has several exact requirements. Give the source material, list the rules the output must satisfy and state the artifact to return. A document-extraction request might require specified fields, original identifiers, a defined missing-value representation and no prose outside the record. A writing task might require a particular audience and section order while keeping supplied conditions unchanged.
The repetition change addresses generation behavior rather than adding a new reasoning mode. For a long answer, define the required sections and the purpose of each so that the request has a clear stopping point. The function-template change concerns how tool-enabled conversation is represented; use the matching edition's conversation and tokenizer settings rather than treating an older formatting example as interchangeable.
Visual and multilingual tasks
The inherited vision feature allows a text question to be paired with an image. Supply a readable page, chart or diagram and identify the detail needed by the task. For a chart comparison, retain the axes, units and legend and request the observation supporting each conclusion. For a document field, keep the label and surrounding context so the returned value can be interpreted correctly.
The documented language coverage includes Russian, English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, Farsi, Hindi, Bengali and other languages listed by the maker. Specify the final language and any terminology that should remain unchanged. A bilingual document task can preserve identifiers and units while adapting the surrounding explanation to the reader.
For a practical visual extraction, ask for the document identifier, field label, extracted text and the part of the image supporting it. An explanation can follow separately if the reader needs interpretation. This keeps transcription, structured fields and the model's conclusion from becoming one indistinguishable result.
System instructions and tools
The 3.2 card recommends a system prompt tailored to the use case and a relatively low temperature, such as 0.15. Set the assistant's role, response rules and treatment of source material explicitly. Place the actual document or code with the user task, preserving the boundary between instructions about the answer and instructions quoted inside the source.
Function calling needs the application to supply tool descriptions and required arguments. The model prepares a call, the application executes the function and returns its result, and the next answer uses that information. The released examples include visual reasoning, tool calls and text-only instructions. A lookup task can therefore ask the model to select a described operation while leaving the actual lookup and its result to the surrounding application.
For a coding request, provide the relevant function, the input exposing the issue and the behavior that should result. Ask for a proposed correction with the original interface preserved, then examples that distinguish the intended behavior. For a mathematical request, define variables and assumptions and ask for the important justified steps and a separate final expression. These are applications of the inherited programming and reasoning capabilities; execution of proposed code remains a separate step.
Context and edition boundaries
Use the 128k window for the source passages, images, conversation and expected answer together. A large-document comparison can retain related sections while returning a short matrix of findings. A large configured image allowance or generation value appearing in a serving example is not an independently stated maximum capacity of the model.
Small 3.2 is an instruction update with documented visual input, not the text-only Small 3 2501 or the hybrid Small 4 reasoning release. The maker's card does not provide a separate hard output-token ceiling for this entry. Preserve its exact 2506 identity and formatting, and define the structured record, explanation or code proposal needed by the next step.
POST /v1/chat/completionstexttext| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits mistralai/mistral-small-3.2-24b-instruct | 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 $0.0002 / 1K tokens Output · Output tokens $0.0006 / 1K tokens Input · Cache read $0.00002 / 1K tokens Input · Cache write — / 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 $0.0002 / 1K tokens Output · Output tokens $0.0006 / 1K tokens Input · Cache read $0.00002 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
set -eu\n: "${AILOOT_API_KEY:?Set your Ailoot API key}"\npython3 - <<'PY_REQUEST'\nimport json,os\nfrom pathlib import Path\np=json.loads('{"model": "mistralai/mistral-small-3.2-24b-instruct", "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-small-3.2-24b-instruct',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-small-3.2-24b-instruct',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-small-3.2-24b-instruct
Use the method paths shown above.
mistralai/mistral-small-3.2-24b-instructMistral Small 3.2 24B Instruct is a text-and-image instruction model with a 128k maker context. Its 2506 update specifically improves precise instruction following, reduces repetition and strengthens the function-calling template while retaining Small 3.1's core features.
This description covers the model’s capabilities. Available in Ailoot inputs, outputs and options are listed for the selected method below.
Mistral Small 3.2 24B Instruct 2506 is a targeted update of Mistral Small 3.1 24B Instruct 2503. The maker identifies three specific changes: following precise instructions more closely, reducing repetitive or runaway answers, and making the function-calling template more robust. The model remains an instruction-tuned text-and-image model that generates text rather than a new audio or image-generation edition.
What changes and what carries forward
The 3.2 card explicitly states that its key features are the same as the named Small 3.1 release. Those features include vision, multilingual work, system-prompt support, native function calling, JSON output and a 128k context window. The referenced 3.1 card describes a 24-billion-parameter model and a Tekken tokenizer with a vocabulary labeled 131k. This documented link supports carrying those features into 3.2 while preserving the 2506 update as its own edition.
The instruction-following focus matters when the result has several exact requirements. Give the source material, list the rules the output must satisfy and state the artifact to return. A document-extraction request might require specified fields, original identifiers, a defined missing-value representation and no prose outside the record. A writing task might require a particular audience and section order while keeping supplied conditions unchanged.
The repetition change addresses generation behavior rather than adding a new reasoning mode. For a long answer, define the required sections and the purpose of each so that the request has a clear stopping point. The function-template change concerns how tool-enabled conversation is represented; use the matching edition's conversation and tokenizer settings rather than treating an older formatting example as interchangeable.
Visual and multilingual tasks
The inherited vision feature allows a text question to be paired with an image. Supply a readable page, chart or diagram and identify the detail needed by the task. For a chart comparison, retain the axes, units and legend and request the observation supporting each conclusion. For a document field, keep the label and surrounding context so the returned value can be interpreted correctly.
The documented language coverage includes Russian, English, French, German, Spanish, Italian, Portuguese, Chinese, Japanese, Korean, Arabic, Farsi, Hindi, Bengali and other languages listed by the maker. Specify the final language and any terminology that should remain unchanged. A bilingual document task can preserve identifiers and units while adapting the surrounding explanation to the reader.
For a practical visual extraction, ask for the document identifier, field label, extracted text and the part of the image supporting it. An explanation can follow separately if the reader needs interpretation. This keeps transcription, structured fields and the model's conclusion from becoming one indistinguishable result.
System instructions and tools
The 3.2 card recommends a system prompt tailored to the use case and a relatively low temperature, such as 0.15. Set the assistant's role, response rules and treatment of source material explicitly. Place the actual document or code with the user task, preserving the boundary between instructions about the answer and instructions quoted inside the source.
Function calling needs the application to supply tool descriptions and required arguments. The model prepares a call, the application executes the function and returns its result, and the next answer uses that information. The released examples include visual reasoning, tool calls and text-only instructions. A lookup task can therefore ask the model to select a described operation while leaving the actual lookup and its result to the surrounding application.
For a coding request, provide the relevant function, the input exposing the issue and the behavior that should result. Ask for a proposed correction with the original interface preserved, then examples that distinguish the intended behavior. For a mathematical request, define variables and assumptions and ask for the important justified steps and a separate final expression. These are applications of the inherited programming and reasoning capabilities; execution of proposed code remains a separate step.
Context and edition boundaries
Use the 128k window for the source passages, images, conversation and expected answer together. A large-document comparison can retain related sections while returning a short matrix of findings. A large configured image allowance or generation value appearing in a serving example is not an independently stated maximum capacity of the model.
Small 3.2 is an instruction update with documented visual input, not the text-only Small 3 2501 or the hybrid Small 4 reasoning release. The maker's card does not provide a separate hard output-token ceiling for this entry. Preserve its exact 2506 identity and formatting, and define the structured record, explanation or code proposal needed by the next step.
POST /v1/messagestextjson, sse| Field | Type | Requirement | Values & limits | Default |
|---|---|---|---|---|
Fieldmodel | Typestring | RequirementRequired | Values & limits mistralai/mistral-small-3.2-24b-instruct | 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 $0.0002 / 1K tokens Output · Output tokens $0.0006 / 1K tokens Input · Cache read $0.00002 / 1K tokens Input · Cache write — / 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 $0.0002 / 1K tokens Output · Output tokens $0.0006 / 1K tokens Input · Cache read $0.00002 / 1K tokens Input · Cache write — / 1K tokens Input · Cache write, 1 hour — / 1K tokens |
Ailoot reserves an estimated amount before a request. After the result, it settles the final charge from measured usage. Streaming settlement uses the final usage record. How reservation and settlement work →
{\n "model": "mistralai/mistral-small-3.2-24b-instruct",\n "max_tokens": 128,\n "messages": [\n {\n "role": "user",\n "content": "Summarize the incident report."\n }\n ],\n "stream": false\n}set -eu\n: "${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-small-3.2-24b-instruct', '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-small-3.2-24b-instruct', '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-small-3.2-24b-instruct
Use the method paths shown above.
Technical example in English
Extract exactly the requested fields\n\nRead this note: 'Service case C-52. Customer: Dana Lee. Device: Lamp Mini. Reported problem: switches off after ten minutes.' Return exactly four labelled lines, in this order: Case ID, Device, Problem, Purchase date. Use unknown for missing information. Omit the customer's name and all introductory or closing prose.Exactly four labelled lines containing source-derived values or unknown, in the requested order.Purchase date is unknown, Dana Lee is omitted, and there are no extra lines.