Mistral AI · Released Oct 6, 2026
Mistral Large 4 Preview
Mistral's reasoning-optimized model featuring verified performance on the OpenRating benchmark index, balanced for high throughput and enterprise reliability.
OpenRating Index
38.0
0–100 composite
Reasoning Index
51.3
0–100 composite
Cost / Task
$1.13
benchmark run
Output speed
116 tok/s
blended tok/s
Time to first token
1.5s
median latency
Input price
$1.36
per 1M tokens
Output price
$4.18
per 1M tokens
Context window
524K
tokens
Strengths
Intelligence Index score of 38.0 on independent evaluations
116.1 tokens/sec median generation throughput with sub-second response times
524k token context window supporting complex multi-document reasoning
How we rate Mistral Large 4 Preview
OpenRating scores every model on the same axes. The OpenRating Index is a weighted composite of reasoning, knowledge, and agentic benchmarks, normalised to a 0–100 scale. Speed is blended output measured in tokens per second under standardised load, with time-to-first-token reported separately. Prices are list input and output prices per million tokens, and Task Cost measures end-to-end workload spending.
OpenRating Index
38.0
Reasoning Index
51.3
Token Pricing
Input per 1M tokens
$1.36
Output per 1M tokens
$4.18
Cost of 1M input + 1M output
$5.54
Task Cost Economics
Empirical cost breakdown for a standard OpenRating Index benchmark task ($1.13 total).
Related models
Compare Mistral Large 4 Preview with the models it is most often measured against.