Skip to content
Model Pareto

Mistral Large 4

Mistral AIProprietarymistral.ai ↗
Compare with nearest rivalsCompare with…
Input
$1.36/M tok
Output
$4.18/M tok
Blended
$2.065/M tok · 3:1
Speed
~100tok/sest.
Context
—
Price
—
Time
—

Preview; weights "drop end of this month" and are not downloadable yet. Launch post states 1T total / 49B active; once weights ship, set openWeights=true, re-add params and verify against the HF card.

Rankings

Benchmark scores

~ italic, dashed = no published score yet, estimated from related benchmarks.

Coding

SWE-bench VerifiedEstimated
~#2 of 14 (estimated)~87.0%
  1. #1Claude Opus 596.0%
  2. ~#2Mistral Large 4Estimated87.0%
  3. #2Claude Sonnet 585.2%
No published score — estimated from related benchmarks
SWE-bench Pro (Public)Estimated
~#6 of 20 (estimated)~68.0%
  1. #1Claude Opus 5.589.9%
  2. #2Claude Sonnet 5.581.3%
  3. #5Intern-S2-397B68.5%
  4. ~#6Mistral Large 4Estimated68.0%
  5. #6Hy4 preview65.7%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#6 of 25 (estimated)~88.5%
  1. #1Qwen3.8-Omni-Flash92.6%
  2. #2Qwen3.8 Flash-Next91.9%
  3. #5Nemotron 3 Ultra89.0%
  4. ~#6Mistral Large 4Estimated88.5%
  5. #6gpt-oss-120b87.8%
No published score — estimated from related benchmarks
SciCodeEstimated
~#27 of 106 (estimated)~53.3%
  1. #1Claude Opus 5.566.9%
  2. #2Claude Fable 5.163.1%
  3. #26DeepSeek-V4-Pro-0813-nvfp4-DSpark53.8%
  4. ~#27Mistral Large 4Estimated53.3%
  5. #27Grok 4.653.0%
No published score — estimated from related benchmarks

Agents

OSWorld-VerifiedEstimated
~#3 of 12 (estimated)~83.4%
  1. #1Kimi K384.8%
  2. #2Qwen3.8 27B84.3%
  3. ~#3Mistral Large 4Estimated83.4%
  4. #3Nex-N2.5-Pro82.2%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~55.6%
  1. #1GLM-5.3 Flash59.1%
  2. #2Kimi K358.3%
  3. ~#3Mistral Large 4Estimated55.6%
  4. #3GPT-5.6 Terra50.2%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#5 of 18 (estimated)~90.5%
  1. #1Atria Dawn Preview92.5%
  2. #2GPT-6 Astra91.5%
  3. #4Claude Opus 590.8%
  4. ~#5Mistral Large 4Estimated90.5%
  5. #5Nex-N2.5-Pro89.7%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#14 of 88 (estimated)~1585 Elo
  1. #1Claude Opus 5.51866
  2. #2Claude Fable 5.11758
  3. #13Qwen3.8 2.4T A95B1597
  4. ~#14Mistral Large 4Estimated1585
  5. #14GPT-6.1 Sol1575
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~#12 of 90 (estimated)~43.2%
  1. #1Muse Spark 1.350.5%
  2. #2GLM-5.350.3%
  3. #11Grok 4.643.3%
  4. ~#12Mistral Large 4Estimated43.2%
  5. #12Claude Opus 542.1%
No published score — estimated from related benchmarks