tensor.news
Integrity IndexModelsCompareLabsPapersMethodologyAbout
tensor.news

© 2026

Evidence

  • Integrity Index
  • Model Claim Ledger
  • Compare

Entities

  • Models
  • Benchmarks
  • Labs
  • Papers

Method

  • Methodology
  • About
  • Privacy

Machine

  • Feed (Atom)
  • llms.txt
  • llms-full.txt
  • Sitemap

tensor.news grades benchmarks and tags model claims by evidence status. Every fact carries a source record and provenance. Scores are framed as task performance under a disclosed harness, never as deployed capability.

← Models

Mixtral 8x22B

Mistral AI·released 2024-04-17

independently reproduced

Scored on 5 benchmarks. 40% of its scores are independently reproduced — the rest are self-reported or unverified.

5
Benchmarks scored
0
SOTA leads
70.4
Best score
40%
Reproduced
2 of 5

Benchmark fingerprint

Every score, ranked, colored by evidence status.

  • MMLU
    70.4
  • MATH level 5
    24.24
  • GPQA diamond
    12.08
  • Cybench
    7.5
  • WeirdML
    3.17

All benchmark scores

BenchmarkScoreEvidence statusMeasured
MMLU70.4accuracy (%)unverified· optimized2024-04-17
MATH level 524.24accuracy (%)reproduced· optimized2024-04-17
GPQA diamond12.08accuracy (%)reproduced· optimized2024-04-17
Cybench7.5accuracy (%)unverified· optimized2024-04-17
WeirdML3.17accuracy (%)unverified2024-04-17
5 source records · epoch_benchmarks. Every score is evidence-linked.

Follow the record

More from Mistral AI

  • Mistral NeMo84.2
  • Mixtral 8x7B83.07
  • Mistral Medium 381.63
  • Mistral 7B v0.175.2
  • Mistral Large 2 (Jul 2024)73.33
  • Mistral Large58.4