Mistral Small 4
- In Input modalities: TextImage Out Output modalities: Text
- Reasoning· none, high
- Reasoning can be off
- Tool use
- Structured output
- Open weights
Prices
1 offering · prices as listed Mistral AI API (opens in a new tab)
mistral-small-2603| Conditions | Input/ 1M tokens | Cache read/ 1M tokens | Output/ 1M tokens | Audio/ 60 seconds |
|---|---|---|---|---|
| Standard | $0.15 | $0.015 | $0.60 | — |
| Voice agent | — | — | — | $0.040 |
Prices from Mistral pricing (opens in a new tab) (checked Nov 4, 2026)
Standard: the price a request with no conditions pays (the headline price elsewhere on the site).
Details
- Knowledge cutoff
- Jun 2025
- Licence
- Apache 2.0
- Family
- mistral-small
- Aliases
- mistral-small-2603
- Model ID
Not stated on Mistral AI’s pages: max input, max thinking budget, built-in tools.
Cited scores
From third-party leaderboards, not measured by usThis model isn’t on any leaderboard we cite yet.
Provider-reported scores
Published by Mistral AI, under the conditions it states| Benchmark | Setting | Score |
|---|---|---|
| GPQA Diamond | reasoning | 71.2% |
| instruct (no reasoning) | 59.1% | |
| MMLU-Pro | reasoning | 78% |
| instruct (no reasoning) | 73.5% | |
| IFBench | reasoning | 48% |
| instruct (no reasoning) | 35.7% | |
| MMMU-Pro | reasoning | 60% |
| instruct (no reasoning) | 46.3% | |
| AIME 2025 | reasoning_effort=high | 83.8% |
| instruct (no reasoning) | 36% | |
| LiveCodeBench | reasoning_effort=high | 63.6% |
| instruct (no reasoning) | 32% |
Changes
All changes →- 1 new price on Mistral AI API Audio (Voice agent) $0.040
- 1 new price on Mistral AI API Cache read $0.015
Sources
- mistralai/Mistral-Small-4-119B-2603 model card (Hugging Face) (opens in a new tab) huggingface.co · checked Sep 26, 2026
- Mistral pricing (opens in a new tab) docs.mistral.ai · checked Nov 4, 2026
- Mistral docs: mistral-small-4-0-26-03 (opens in a new tab) docs.mistral.ai · checked Nov 3, 2026
- Mistral Small 4 model card (mistralai on Hugging Face) (opens in a new tab) huggingface.co · checked Nov 8, 2026
- models.dev: mistral-small-2603 (community data, MIT) (opens in a new tab) github.com · checked Sep 30, 2026
- Introducing Mistral Small 4 | Mistral AI (opens in a new tab) mistral.ai · checked Nov 4, 2026
Mistral AI’s official policies: data use, privacy, terms →
Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data
Mistral AI · France · mistral.ai
Related models
Snapshots, the same family and other models from this provider
- Mistral Large 4
- Ministral 3 14B
- Ministral 3 3B
- Ministral 3 8B
- Magistral Medium 1.2
- Mistral Nemo 12B
- Mistral-7B-Instruct-v0.1
- Mistral-Large-4-1T-A52B
- Claude Haiku 5.5 vs Mistral Small 4
- Gemini 3.5 Flash-Lite vs Mistral Small 4
- GLM-5.3-Flash vs Mistral Small 4
- GPT-5.4 Mini vs Mistral Small 4
- GPT-6 Luna vs Mistral Small 4
- Mercury 2.5 vs Mistral Small 4
- MiMo-V2.6-Flash vs Mistral Small 4
- Mistral Small 4 vs Qwen3.8-Flash