Providers and ModelsMistral AI

Mistral Small 4

Mistral Small 4
Mistral AI · Text Generation
POST /v1/chat/completions

Hybrid model unifying Instruct, Reasoning (Magistral), and Devstral families: 40% lower completion time and 3x throughput vs Small 3.

At a glance

FieldValue
Model idmistral-small-4
Model release date2026-03-16
Input modalitiesText, Image
Output modalitiesText
Context window256K
Weight precision-
Max output tokens65,536
Featuresvision, web_search, function_calling
Native inferenceNo
NewNo
Structured outputJSON Mode
Batch API35% off list price
Supported endpointsPOST /v1/chat/completions, POST /v1/responses, POST /v1/messages, POST /v1beta/models/mistral-small-4:generateContent
Alternate model idsmistralai/mistral-small-4

Pricing

ChargeSpecRate
Inputper 1M prompt tokens$0.15
Outputper 1M generated tokens$0.60
Standard Web Searchper call$0.084
Premium Web Searchper call$0.140
Code Interpreterper call$0.084
Image Generationper image$0.280

Example request

$curl https://api.empiriolabs.ai/v1/chat/completions \
>  -H 'Authorization: Bearer $EMPIRIOLABS_API_KEY' \
>  -H 'Content-Type: application/json' \
>  -d '{"model": "mistral-small-4", "messages": [{"role":"user","content":"Hello"}]}'

Parameters

ParameterTypeRequiredDefaultDescription
reasoning_enabledbooleannotrueEnable extended reasoning (maps to reasoning: high|none)
tool_web_searchbooleannotrueEnable web_search tool
web_search_tierenumno"standard"Standard or Premium web-search tier. Premium uses higher-quality sources. · Allowed: standard, premium
tool_code_interpreterbooleannotrueAllow the model to execute Python code in a sandbox to compute / analyze data.
tool_image_generationbooleannotrueAllow the model to generate images inline via the platform image-gen tool.
temperaturenumberno0.7Sampling temperature. 0 = deterministic, 2 = maximum randomness. · Range: 0 – 1.5
max_tokensnumberno4096Maximum tokens in the response. · Range: 1 – 32768
disable_formattingbooleannofalseSkip the EmpirioLabs Markdown formatting (citation [N] rewriting + References block when the web_search tool was used). The raw upstream answer with plain [N] citations is returned.
response_formatenumno-Return the output as a valid JSON object (JSON mode). Describe the fields you want in your prompt.

Notes

Tools (web search, code interpreter, image generation) are billed only when actually invoked. Requests that include your own function tools use standard function calling, and the built-in tools are unavailable on those requests: a request cannot combine both at once.

Per-tool billing (usage.tool_usage)

When this model invokes built-in tools (web search, code interpreter, etc.) inside a single request, the response carries a normalized usage.tool_usage map alongside the token counts. The example below shows the shape — exact field names, units, and which tools appear can vary slightly per provider:

1"usage": {
2  "prompt_tokens": 123,
3  "completion_tokens": 456,
4  "cost_usd": 0.0042,
5  "tool_usage": {"web_search": 3, "code_interpreter": 1}
6}

The tool counts are already factored into cost_usd — they are surfaced for transparency so you can audit per-tool billing. The field is omitted when no tools were invoked.

Machine-readable schema: GET https://api.empiriolabs.ai/v1/models/mistral-small-4.