LLM Perks

Mistral Small

Mistral AI · All Mistral models
Free right now Mid-size
Tool useLong contextOpen weights

Mistral Small is a language model with 24B parameters from Mistral AI. It is currently free at 2 providers with up to 262K tokens of context. The model supports tool calling. Its weights are open, so it can also run locally.

Developer
Mistral AI
Parameters
24B
Context
262K tokens
Type
Text
License
Open weights
Free since
September 19, 2026

Where to use Mistral Small for free

Each row is a separate free endpoint with its own limits.

ProviderContextLimitCard
Mistral AI
mistral-small-latest
33K 1 req/min · 500 tokens/min — Open
UnoRouter
mistral-small:free
262K No-card free models use shared capacity and per-model limits; availability can change. No Open

Quick start

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

Example for Mistral AI. Any OpenAI-compatible client works: Cursor, Cline, OpenCode, Open WebUI.

Benchmarks

Intelligence Index
5.8
GPQA Diamond
38.1%
Humanity's Last Exam
4.3%

Source: Artificial Analysis · Mistral Small (Sep '24)

Similar free models

FAQ

Is Mistral Small free?

Yes. As of October 24, 2026, Mistral Small is free at Mistral AI, UnoRouter. Each provider has its own limits — see the table above.

What is the context window of Mistral Small?

The largest context among free routes is 262,100 tokens. Some providers cap it lower.

How do I call Mistral Small via API?

Use any OpenAI-compatible client with base URL https://api.mistral.ai/v1, model ID mistral-small-latest, and a Mistral AI key.

How good is Mistral Small?

Artificial Analysis Intelligence Index: 5.8, GPQA Diamond: 38%. For comparison, the strongest free model in the catalog right now is GLM 5.3.

Is Mistral Small open-weight?

Yes, the weights are published — you can download and run it locally, for example with Ollama or LM Studio.