Benchmark menu

Model detail

Mistral Medium 3.5

A static, shareable model page that remains complete without JavaScript.

Mistral Medium 3.5

Provider: Mistral · Public slug: mistralai-mistral-medium-3-5

Rank
34
Rank movement
Baseline
Score
15.3
Last measured
2026-07-04
Coverage
Medium not supported

Reasoning scores

XHigh
15.1
Medium
None
15.5

Generation health

XHigh22%22%56%n 9None25%25%50%n 8 Medium not supportedHigh failure rate

Cost summary

Mean standard list-price equivalent
$0.12
Median unique-cell cost
$0.07
Unique-cell range
$0.02–$0.39
Coverage
16 unique cells / 17 observed runs
Observed-run total
$2.26
Evidence basis
Provider-reported
Mean output
15k tok
Output range
1.5k–53k tok
Token samples
17

Cost means equal-weight unique model/game/reasoning cells. Provider-account billed spend is reported separately on Charts.

Run setup

  • Overall Models: combines public reasoning variants
  • Standard GameBench code-generation prompt