SmophyAI

Published daily · Next update 02:00 UTC

Model family · 16 versions tracked

Mistral: every version compared

Mistral: Mistral Small 4 (batch) currently leads the Mistral family with an Artificial Analysis Intelligence Index of 11.3. For the best value within the family, Mistral: Ministral 3 3B 2512 costs $0.10/1M tokens blended.

  1. Mistral: Mistral Small 4 (batch) - 11.3 intelligence · $0.26/1M
  2. Mistral: Mistral Large 3 2512 (batch) - 9.3 intelligence · $0.75/1M
  3. Mistral: Mistral Medium 3.1 (batch) - 9.2 intelligence · $0.80/1M
  4. Mistral: Devstral 2 2512 - 8.6 intelligence · $0.80/1M
  5. Mistral: Ministral 3 14B 2512 - 6 intelligence · $0.20/1M
  6. Mistral: Ministral 3 8B 2512 (batch) - 5.5 intelligence · $0.15/1M
  7. Mistral: Ministral 3 3B 2512 - 4.8 intelligence · $0.10/1M
  8. Mistral: Mistral Nemo - - intelligence · $0.02/1M
  9. Mistral: Voxtral Small 24B 2507 - - intelligence · $0.15/1M
  10. Mistral Large 2407 - - intelligence · $3.00/1M
  11. Mistral: Codestral 2508 (batch) - - intelligence · $0.45/1M
  12. Mistral: Mistral Medium 3 - - intelligence · $0.80/1M
  13. Mistral: Mistral Small 3 - - intelligence · $0.06/1M
  14. Mistral: Mixtral 8x22B Instruct - - intelligence · $3.00/1M
  15. Mistral Large - - intelligence · $3.00/1M
  16. Mistral: Ministral 8B - - intelligence · $0.11/1M
All 16 Mistral versions, by quality
#ModelIntelligence indexPrice/1M blended
01Mistral: Mistral Small 4 (batch)11.3$0.26
02Mistral: Mistral Large 3 2512 (batch)9.3$0.75
03Mistral: Mistral Medium 3.1 (batch)9.2$0.80
04Mistral: Devstral 2 25128.6$0.80
05Mistral: Ministral 3 14B 25126$0.20
06Mistral: Ministral 3 8B 2512 (batch)5.5$0.15
07Mistral: Ministral 3 3B 25124.8$0.10
08Mistral: Mistral Nemo-$0.02
09Mistral: Voxtral Small 24B 2507-$0.15
10Mistral Large 2407-$3.00
11Mistral: Codestral 2508 (batch)-$0.45
12Mistral: Mistral Medium 3-$0.80
13Mistral: Mistral Small 3-$0.06
14Mistral: Mixtral 8x22B Instruct-$3.00
15Mistral Large-$3.00
16Mistral: Ministral 8B-$0.11

Mistral real usage share, over time

0.7%0.5%0.3%0.2%0.0%Share of usageJul 5Jul 28Sep 20

Ranked by current quality, not release date - this pipeline's data collection started recently, so "first tracked" timestamps don't reflect true historical release dates yet. The usage-share chart above is real, unaffected by that limitation.

Frequently asked questions

What is the best Mistral model?

Mistral: Mistral Small 4 (batch) currently leads the Mistral family with an Artificial Analysis Intelligence Index of 11.3.

How much real usage does Mistral get?

Mistral models account for 0.3% of real OpenRouter token volume as of the latest data.