M
Mistral Small 4OPEN
Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system.
Prices update automatically — checked daily against provider list prices.
- $0.26 per 1M tokens, assuming 3 input for 1 output the convention used by Artificial Analysis
- cheaper than 57% of the 191 open-weight models tracked here
- 8 of 19 Mistral AI models tracked here, from cheapest
Benchmarks & performance
Independent benchmark scores for Mistral Small 4, measured by Artificial Analysis. Higher is better. Measurement mode: Reasoning.
For the same budget, Mistral Small 4 delivers 12 times more capability. Open weights also mean you can self-host it and pay nothing per token.
Models at this level cost $0.91 per 1M tokens (median of 23) — this one costs $0.26.
Cheaper and better on this index: GLM 5.3 Flash · GPT-5.6 Luna (batch) · DeepSeek V4 Flash 0731 and 16 more
About this model
Mistral Small 4 is an open-weight AI model by Mistral AI. You can download and self-host it for free; the prices below are hosted-API list prices, tracked daily, for when you prefer convenience over self-hosting.
Frequently asked questions
What is Mistral Small 4?
Mistral Small 4 is an AI language model from Mistral AI. It is open-weight: you can download it and run it on your own hardware, for free. It scores 19.7 on the Artificial Analysis intelligence index.
Is Mistral Small 4 free?
The weights are free and open — you can self-host Mistral Small 4 and pay nothing per token. If you prefer a hosted API, list prices are $0.15 per million input tokens and $0.6 per million output tokens.
What is Mistral Small 4 good at?
Independent benchmarks from Artificial Analysis give it GPQA 76.9%, Humanity's Last Exam 9.9%, Long Context Reasoning 47.3%, SciCode 38%, IFBench 48.2%, τ²-Bench 41.2%, τ-Bench Banking 4.9%, Terminal-Bench 21%, Terminal-Bench Hard 17.4%. It is particularly used for code generation.
How fast is Mistral Small 4?
It generates about 164.4 tokens per second, with a median 0.67s delay before the first token. Measured independently by Artificial Analysis.
Can I self-host Mistral Small 4?
Yes. Mistral Small 4 has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.