Mistral AI model product
Mistral Small 4
Mistral Small 4 is Mistral AI's hybrid small model that unifies instruct, reasoning, multimodal understanding, and coding capabilities in a single efficient system.
Updated Aug 12, 2026. Default version: Mistral Small 4
LLMBoard score33.4Mistral Small 4
Coverage100%9 benchmark families
Context window256KTokens
Official input price$0.15Mistral API
Mistral Small 4 Specifications
Technical details for the model's default version.
- Version
- Mistral Small 4
- Released
- Mar 16, 2026
- Knowledge cutoff
- Jun 1, 2025
- Parameters
- 119B
- Context window
- 256K
- Max output
- 256K
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Apache 2.0
Mistral Small 4 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Mistral Small 4 Benchmark Results
Benchmark scores for Mistral Small 4.
| AA-LCR | 71.2% | 03 | 16 | 86.7% | C | |
| COLLIE | 62.9% | 07 | 10 | 33.3% | C | |
| Arena Hard | 58.3% | 13 | 26 | 52.0% | C | |
| IFBench | 48.0% | 26 | 29 | 10.7% | C | |
| LiveCodeBench | 63.6% | 29 | 73 | 61.1% | C | |
| MMMU-Pro | 60.0% | 50 | 66 | 24.6% | C | |
| MMLU-Pro | 78.0% | 60 | 129 | 53.9% | C | |
| AIME 2025 | 83.8% | 62 | 114 | 46.0% | C | |
| GPQA | 71.2% | 125 | 234 | 46.8% | C |
Mistral Small 4 Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
Mistral Small 4 Runtime Performance
Provider-specific output speed and catalog latency for Mistral Small 4. Runtime does not affect the capability score.
| Mistral AI | 8.772 tok/s | N/A | 256K | 256K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Mistral Small 4 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.15 input, $0.60 output per 1M
- Official provider
- Mistral
- Lowest third-party
- From $0.15 input, $0.60 output per 1M via Merge Gateway
- Tracked offerings
- 3
| Mistral | mistral-small-latest | global | $0.15 | $0.60 | 256K | |
| Mistral | mistral-small-2603 | global | $0.15 | $0.60 | 256K | |
| Merge Gateway | mistral/mistral-small-latest | global | $0.15 | $0.60 | 256K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Mistral Small 4 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Mistral Small 4 | 33.4 | 119B | 256K | 256K | Yes | Apache 2.0 |
What is Mistral Small 4?
Key information about Mistral Small 4 and its available data.
Mistral Small 4 is Mistral AI's hybrid small model that unifies instruct, reasoning, multimodal understanding, and coding capabilities in a single efficient system.
It uses a 119B-parameter Mixture-of-Experts architecture with native image input, a 256k context window, and configurable reasoning depth for general chat, coding agents, and long-context reasoning workloads.
Data as of 2026-08-11.
Mistral Small 4 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Mistral Small 4
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MA
Ministral 3
Mistral AI
MA
Devstral Medium
Mistral AI
MA
Pixtral Large
Mistral AI
MA
Min istral 3
Mistral AI
MA
Magistral Small
Mistral AI
MA
Magistral Medium
Mistral AI
FAQ
Common questions about Mistral Small 4.
When was Mistral Small 4 released?
Mistral Small 4's default version was released on Mar 16, 2026.
How much does Mistral Small 4 cost?
Mistral Small 4's official API price is $0.15 per million input tokens and $0.60 per million output tokens via Mistral. The lowest tracked third-party offer starts at $0.15 input and $0.60 output via Merge Gateway.
Who created Mistral Small 4?
Mistral Small 4 was created by Mistral AI.
What is the context window for Mistral Small 4?
The default version has a 256K token context window.
Is Mistral Small 4 open weight?
Yes. The default version is marked as open weight under Apache 2.0.
How many API providers offer Mistral Small 4?
3 provider offerings are linked to the default version.
What models should I compare Mistral Small 4 with?
Nearby ranked alternatives include Sarvam 30B, o1 pro, DiffusionGemma 26B A4B.