Pixtral Large
Multimodal
Zero-eval
#1VQAv2
#1MM-MT-Bench
#3AI2D
by Mistral AI
About
Pixtral Large is a multimodal language model developed by Mistral AI. This model demonstrates exceptional performance with an average score of 80.5% across 7 benchmarks. It excels particularly in AI2D (93.8%), DocVQA (93.3%), ChartQA (88.1%). The model shows particular specialization in general tasks with an average performance of 91.0%. It supports a 256K token context window for handling large documents. The model is available through 1 API provider. As a multimodal model, it can process and understand text, images, and other input formats seamlessly. Released in 2024, it represents Mistral AI's latest advancement in AI technology.
Pricing Range
Input (per 1M)$2.00 -$2.00
Output (per 1M)$6.00 -$6.00
Providers1
Timeline
AnnouncedNov 18, 2024
ReleasedNov 18, 2024
Specifications
Capabilities
Multimodal
License & Family
License
Mistral Research License (MRL) for research; Mistral Commercial License for commercial use
Base ModelMistral Large 2
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
7 benchmarks
Average Score
80.5%
Best Score
93.8%
High Performers (80%+)
4Performance Metrics
Max Context Window
256.0KAvg Throughput
0.1 tok/sAvg Latency
1msTop Categories
general
91.0%
vision
79.4%
roleplay
74.0%
math
69.4%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark
AI2D
Rank #3 of 17
#1GPT-4o
94.2%
#2Claude 3.5 Sonnet
94.7%
#3Pixtral Large
93.8%
#4Mistral Small 3.2 24B Instruct
92.9%
#5Llama 3.2 90B Instruct
92.3%
#6Llama 3.2 11B Instruct
91.1%
DocVQA
Rank #12 of 26
#9DeepSeek VL2
93.3%
#10Nova Pro
93.5%
#11Grok-2
93.6%
#12Pixtral Large
93.3%
#13Grok-2 mini
93.2%
#14Phi-4-multimodal-instruct
93.2%
#15GPT-4o
92.8%
ChartQA
Rank #7 of 24
#4Qwen2-VL-72B-Instruct
88.3%
#5Llama 4 Scout
88.8%
#6Nova Pro
89.2%
#7Pixtral Large
88.1%
#8Mistral Small 3.2 24B Instruct
87.4%
#9Qwen2.5 VL 7B Instruct
87.3%
#10Nova Lite
86.8%
VQAv2
Rank #1 of 3
#1Pixtral Large
80.9%
#2Pixtral-12B
78.6%
#3Llama 3.2 90B Instruct
78.1%
MM-MT-Bench
Rank #1 of 3
#1Pixtral Large
74.0%
#2Pixtral-12B
60.5%
#3Qwen2.5-Omni-7B
6.0%
All Benchmark Results for Pixtral Large
Complete list of benchmark scores with detailed information
AI2D AI2D benchmark | general | text | 0.94 | 93.8% | Self-reported |
DocVQA DocVQA benchmark | vision | multimodal | 0.93 | 93.3% | Self-reported |
ChartQA ChartQA benchmark | general | multimodal | 0.88 | 88.1% | Self-reported |
VQAv2 VQAv2 benchmark | vision | multimodal | 0.81 | 80.9% | Self-reported |
MM-MT-Bench MM-MT-Bench benchmark | roleplay | text | 74.00 | 74.0% | Self-reported |
MathVista MathVista benchmark | math | text | 0.69 | 69.4% | Self-reported |
MMMU MMMU benchmark | vision | multimodal | 0.64 | 64.0% | Self-reported |