Mistral AI model product
Pixtral Large
Pixtral Large is a 124B-parameter multimodal model from Mistral AI for understanding text, documents, charts, and natural images.
Updated Sep 8, 2026. Default version: Pixtral Large
LLMBoard Score20.0Pixtral Large
Coverage0%7 benchmark families
Context window128KTokens
Official input priceN/AOfficial price unavailable
Pixtral Large Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Pixtral Large Benchmark Results
Benchmark scores for Pixtral Large.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkVQAv2 | Score80.90% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkAI2D | Score93.80% | Rank04 | Participants34 | Percentile90.91% | EvidenceC | Evaluated |
| BenchmarkChartQA | Score88.10% | Rank07 | Participants26 | Percentile76.00% | EvidenceC | Evaluated |
| BenchmarkDocVQA | Score93.30% | Rank12 | Participants28 | Percentile59.26% | EvidenceC | Evaluated |
| BenchmarkMM-MT-Bench | Score0.74 points | Rank12 | Participants17 | Percentile31.25% | EvidenceC | Evaluated |
| BenchmarkMathVista | Score69.40% | Rank15 | Participants39 | Percentile63.16% | EvidenceC | Evaluated |
| BenchmarkMMMU | Score64.00% | Rank40 | Participants64 | Percentile38.10% | EvidenceC | Evaluated |
| BenchmarkLM Arena Vision | Score1,089.40 rating | Rank92 | Participants107 | Percentile14.15% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,095.19 rating | Rank94 | Participants107 | Percentile12.26% | EvidenceA | Evaluated |
Pixtral Large Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision | Categorychinese | Rank81 | Rating / score1,081.88 | Votes156 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank82 | Rating / score1,097.83 | Votes156 | ObservationsN/A | Result date |
| Arenavision | Categoryenglish | Rank91 | Rating / score1,099.79 | Votes2,921 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank92 | Rating / score1,089.40 | Votes5,423 | ObservationsN/A | Result date |
| Arenavision style control | Categoryenglish | Rank94 | Rating / score1,099.24 | Votes2,921 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank94 | Rating / score1,095.19 | Votes5,423 | ObservationsN/A | Result date |
Pixtral Large Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Pixtral Large Runtime Performance
Provider-specific output speed and catalog latency for Pixtral Large. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderMistral AI | Output Speed0.10 tok/s | Catalog Latency0.50 s | Max Input128K | Max Output128K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Pixtral Large Specifications
Technical details for the model's default version.
- Version
- Pixtral Large
- Released
- Nov 18, 2024
- Knowledge cutoff
- Unknown
- Parameters
- 124B
- Context window
- 128K
- Max output
- 128K
- Inputs
- image, text
- Outputs
- text
- Open weights
- Yes
- License
- Mistral Research License (MRL) for research; Mistral Commercial License for commercial use
Pixtral Large 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 Large | Released | LLMBoard20.03 | Parameters124B | Context128K | Max output128K | Open weightsYes | LicenseMistral Research License (MRL) for research; Mistral Commercial License for commercial use |
Models similar to Pixtral Large
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MA
Min istral 3
Mistral AI
MA
Magistral Small
Mistral AI
MA
Magistral Medium
Mistral AI
MA
Devstral Medium
Mistral AI
MA
Devstral Small 1.1
Mistral AI
MA
Mistral Large 3
Mistral AI
What is Pixtral Large?
Key information about Pixtral Large and its available data.
Pixtral Large is a multimodal model built on Mistral Large 2, with a 123B-parameter multimodal decoder and a 1B-parameter vision encoder. It has a 128K context window supporting up to 30 high-resolution images.
Data as of 2026-09-08.
FAQ
Common questions about Pixtral Large.
When was Pixtral Large released?
Pixtral Large's default version was released on Nov 18, 2024.
How much does Pixtral Large cost?
No official standard PAYG price is currently available for Pixtral Large.
Who created Pixtral Large?
Pixtral Large was created by Mistral AI.
What is the context window for Pixtral Large?
The default version has a 128K token context window.
Is Pixtral Large open weight?
Yes. The default version is marked as open weight under Mistral Research License (MRL) for research; Mistral Commercial License for commercial use.
How many API providers offer Pixtral Large?
No provider offering is currently linked to the default version.