- Intelligence
- 23.1
- Rank 204 of 768 measured models
- Price/1M
- $1.05
- Rank 418 of 726 measured models
- Context window
- 1M
- Rank 90 of 1296 measured models
- Cost per task
- $1.41
- Rank 151 of 218 measured models
Mistral Large 4 (Medium) scores 23.1 on the Artificial Analysis Intelligence Index, ranking 204th of 768 measured models. The closest models on this index are DeepSeek V4 Pro 0813 (23.1), GPT-5 (23.0), and Qwen3.5 27B (22.9). At $1.05 per million tokens (3:1 input/output blended price), it is cheaper than DeepSeek V4 Pro 0813 ($1.98) and GPT-5 ($3.44), more expensive than Qwen3.5 27B ($0.825).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- DeepSeek V4 Pro 0813
DeepSeek · $1.98 · 72.7 t/s
23.1$1.9872.7 t/sCompare - GPT-5
OpenAI · $3.44 · 98.2 t/s
23.0−0.1$3.4498.2 t/sCompare - Qwen3.5 27B
Alibaba · $0.825 · 75.3 t/s
22.9−0.2$0.82575.3 t/sCompare - Claude 4.1 Opus
Anthropic · $30 · —
22.8−0.3$30—Compare
