Pixtral Large
Intelligence
14
#366/523
Coding
—
Math
2.3
#260/265
Speed
57 tok/s
TTFT: 455.00s
Pricing
$2.00 / $6.00
per 1M tokens (in/out)
A 124B parameter multimodal model built on top of Mistral Large 2, featuring frontier-level image understanding capabilities. Excels at understanding documents, charts, and natural images while maintaining strong text-only performance. Features a 123B multimodal decoder and 1B parameter vision encoder with a 128K context window supporting up to 30 high-resolution images.
Pixtral Large is Mistral’s model designed for various AI applications. It processes at 57.497 tokens per second and is priced at $2 per million input tokens and $6 per million output tokens, targeting professional users.
When to Use Pixtral Large
✓ Best For
- Natural language processing tasks.
- Data analysis and interpretation.
- Content generation.
✗ Not Ideal For
- High-speed real-time applications.
- Complex mathematical computations.
How Pixtral Large Compares
Intelligence Index · Higher is better
Allen Institute for AIInclusionAIMistralxAI
Benchmark Profile
Output Speed · tok/s
OpenAIAlibabaMistralMeta
Math Index
GoogleMetaMistralAI21 Labs
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (13)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 14 | |
| Math Index | 2.3 | |
| MMLU-Pro | 701% | |
| GPQA | 505% | |
| LiveCodeBench | 261% | |
| HLE | 36% | |
| SciCode | 29.2% | |
| IFBench | 34.5% | |
| LCR | 10.3% | |
| Tau2 | 36.5% | |
| AIME | 7% | |
| AIME 2025 | 2.3% | |
| MATH 500 | 71.4% |
Data: Artificial Analysis · Updated: April 10, 2026
Frequently Asked Questions (15)
When was Pixtral Large released?
Pixtral Large was released on November 18, 2024.
Who created Pixtral Large?
Pixtral Large was created by Mistral.
How intelligent is Pixtral Large?
Pixtral Large scores 14 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other open weight non-reasoning models of similar size (median: 13).
How fast is Pixtral Large?
Pixtral Large generates output at 52.6 tokens per second (based on Mistral's API), which is below average compared to other open weight non-reasoning models of similar size (median: 60.6 t/s).
What is the latency of Pixtral Large?
Pixtral Large has a time to first token (TTFT) of 1.49s (based on Mistral's API), which is better than average compared to other open weight non-reasoning models of similar size (median: 1.90s).
How much does Pixtral Large cost?
Pixtral Large costs $2.00 per 1M input tokens (at the higher end, median: $0.54) and $6.00 per 1M output tokens (at the higher end, median: $0.90), based on Mistral's API.
What is Pixtral Large API pricing?
Pixtral Large costs $2.00 per 1M input tokens and $6.00 per 1M output tokens (based on Mistral's API). For a blended rate (3:1 input to output ratio), this is $3.00 per 1M tokens. Pricing may vary by provider.
Is Pixtral Large a reasoning model?
No, Pixtral Large is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
What input modalities does Pixtral Large support?
Pixtral Large supports image input.
What output modalities does Pixtral Large support?
Pixtral Large supports text only output.
Can Pixtral Large process images?
Yes, Pixtral Large supports image input and can analyze, describe, and answer questions about images.
Is Pixtral Large multimodal?
No, Pixtral Large is not multimodal. It only supports image input.
What is the context window of Pixtral Large?
Pixtral Large has a context window of 130k tokens. This determines how much text and conversation history the model can process in a single request.
Is Pixtral Large open source?
Yes, Pixtral Large is open weights. The model weights are publicly available and can be downloaded for self-hosting.
How many parameters does Pixtral Large have?
Pixtral Large has 124 billion parameters.
