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

Mistral Medium 3

Mistral AI·released 2025-05-07

independently reproduced

Scored on 5 benchmarks. 80% of its scores are independently reproduced — the rest are self-reported or unverified.

5
Benchmarks scored
0
SOTA leads
81.63
Best score
80%
Reproduced
4 of 5

Benchmark fingerprint

Every score, ranked, colored by evidence status.

  • MATH level 5
    81.63
  • GPQA diamond
    46.04
  • OTIS Mock AIME 2024-2025
    32.15
  • FrontierMath-2025-02-28-Private
    0.61
  • HLE
    0

All benchmark scores

BenchmarkScoreEvidence statusMeasured
MATH level 581.63accuracy (%)reproduced· optimized2025-05-07
GPQA diamond46.04accuracy (%)reproduced· optimized2025-05-07
OTIS Mock AIME 2024-202532.15accuracy (%)reproduced· optimized2025-05-07
FrontierMath-2025-02-28-Private0.61accuracy (%)reproduced· optimized2025-05-07
HLE0accuracy (%)unverified2025-05-07
5 source records · epoch_benchmarks. Every score is evidence-linked.

Follow the record

More from Mistral AI

  • Mistral NeMo84.2
  • Mixtral 8x7B83.07
  • Mistral 7B v0.175.2
  • Mistral Large 2 (Jul 2024)73.33
  • Mixtral 8x22B70.4
  • Mistral Large58.4