Mistral AI model product
Pixtral 12B
A 12B parameter multimodal model with a 400M parameter vision encoder, capable of understanding both natural images and documents.
Updated Aug 17, 2026. Default version: Pixtral-12B
LLMBoard Score0.0Pixtral-12B
Coverage80%11 benchmark families
Context window128KTokens
Official input price$0.15Mistral API
Pixtral 12B Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Pixtral 12B Benchmark Results
Benchmark scores for Pixtral-12B.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkMM IF-Eval | Score52.7% | Rank02 | Participants2 | Percentile0.0% | EvidenceC | Evaluated |
| BenchmarkVQAv2 | Score78.6% | Rank02 | Participants3 | Percentile50.0% | EvidenceC | Evaluated |
| BenchmarkMM-MT-Bench | Score0.605 points | Rank04 | Participants17 | Percentile81.3% | EvidenceC | Evaluated |
| BenchmarkMT-Bench | Score76.8% | Rank11 | Participants12 | Percentile9.1% | EvidenceC | Evaluated |
| BenchmarkChartQA | Score81.8% | Rank18 | Participants26 | Percentile32.0% | EvidenceC | Evaluated |
| BenchmarkDocVQA | Score90.7% | Rank20 | Participants28 | Percentile29.6% | EvidenceC | Evaluated |
| BenchmarkMathVista | Score58.0% | Rank28 | Participants39 | Percentile28.9% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score72.0% | Rank55 | Participants66 | Percentile16.9% | EvidenceC | Evaluated |
| BenchmarkMMMU | Score52.5% | Rank56 | Participants63 | Percentile11.3% | EvidenceC | Evaluated |
| BenchmarkMATH | Score48.1% | Rank58 | Participants71 | Percentile18.6% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score61.3% | Rank66 | Participants67 | Percentile1.5% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score69.2% | Rank86 | Participants101 | Percentile15.0% | EvidenceC | Evaluated |
Pixtral 12B Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision | Categoryoverall | Rank124 | Rating / score1008.7 | Votes7,511 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank128 | Rating / score1025.6 | Votes7,511 | ObservationsN/A | Result date |
Pixtral 12B Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.15 input, $0.15 output per 1M
- Official provider
- Mistral
- Lowest third-party
- From $0.20 input, $0.20 output per 1M via Scaleway
- Tracked offerings
- 4
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderMistral | Provider model IDpixtral-12b | Regionglobal | Input / 1M$0.15 | Output / 1M$0.15 | Context128K | Updated |
| ProviderScaleway | Provider model IDpixtral-12b-2409 | Regionglobal | Input / 1M$0.20 | Output / 1M$0.20 | Context128K | Updated |
| ProviderCortecs | Provider model IDpixtral-12b-2409 | Regionglobal | Input / 1M$0.223 | Output / 1M$0.223 | Context128K | Updated |
| ProviderGreenPT | Provider model IDpixtral-12b-2409 | Regionglobal | Input / 1M$0.285 | Output / 1M$0.285 | Context128K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Pixtral 12B Runtime Performance
Provider-specific output speed and catalog latency for Pixtral-12B. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderMistral AI | Output Speed0.1 tok/s | Catalog Latency0.5 s | Max Input128K | Max Output8.2K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Pixtral 12B Specifications
Technical details for the model's default version.
- Version
- Pixtral-12B
- Released
- Sep 17, 2024
- Knowledge cutoff
- Sep 1, 2024
- Parameters
- 12.4B
- Context window
- 128K
- Max output
- 128K
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Apache 2.0
Pixtral 12B Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionPixtral-12B | Released | LLMBoard0.0 | Parameters12.4B | Context128K | Max output128K | Open weightsYes | LicenseApache 2.0 |
Pixtral 12B vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Pixtral 12B
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MA
Mistral NeMo
Mistral AI
MA
Shieldstral 1.0
Mistral AI
MA
Ministral 8B
Mistral AI
MA
Codestral 22B
Mistral AI
MA
Mistral Large 2
Mistral AI
MA
Mistral Small 3 24B
Mistral AI
What is Pixtral 12B?
Key information about Pixtral 12B and its available data.
A 12B parameter multimodal model with a 400M parameter vision encoder, capable of understanding both natural images and documents. Excels at multimodal tasks while maintaining strong text-only performance. Supports variable image sizes and multiple images in context.
Data as of 2026-08-17.
FAQ
Common questions about Pixtral 12B.
When was Pixtral 12B released?
Pixtral 12B's default version was released on Sep 17, 2024.
How much does Pixtral 12B cost?
Pixtral 12B's official API price is $0.15 per million input tokens and $0.15 per million output tokens via Mistral. The lowest tracked third-party offer starts at $0.20 input and $0.20 output via Scaleway.
Who created Pixtral 12B?
Pixtral 12B was created by Mistral AI.
What is the context window for Pixtral 12B?
The default version has a 128K token context window.
Is Pixtral 12B open weight?
Yes. The default version is marked as open weight under Apache 2.0.
How many API providers offer Pixtral 12B?
4 provider offerings are linked to the default version.
What models should I compare Pixtral 12B with?
Nearby ranked alternatives include Phi 3.5 mini, Gemini 1.5 Flash 8B, Qwen3.5 0.8B.