Mistral AI·released 2026-04-281 source
Mistral Medium 3.5 benchmark scores: 6 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 6 benchmarks·best result 59.12 on DTBench (unverified)·0 of 6 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| DTBench | 59.12accuracy (%) | unverifiedT1 | 2026-04-28https://conceptualreasoning.ai/dtbench |
| WeirdML | 43.71accuracy (%) | unverifiedT1 | 2026-04-28https://htihle.github.io/weirdml.html |
| LMCA | 30.66accuracy (%) | unverifiedT1 | 2026-04-28https://conceptualreasoning.ai/lmca |
| Surface Evolver Bench | 26.88accuracy (%) | unverifiedT1 | 2026-04-28https://yhenon.github.io/surface-evolver-llm-eval/ |
| ProofBench | 9accuracy (%) | unverifiedT1 | 2026-04-28https://www.vals.ai/benchmarks/proof_bench |
| FrontierCode | 8accuracy (%) | unverifiedT1 | 2026-04-28https://cognition.com/frontiercode |
Claims drawn from cited facts, not live model generation.
This model comes from Mistral AI, placing its origin with a European developer known for releasing openly available weights. Its vintage is 2026, making it a recent arrival on the current generation of model releases.
2 cited facts
This model is scored on 6 tracked benchmarks. It holds no current top score on any of them. Its record remains unverified, meaning outside evaluation has neither confirmed nor disputed the reported results.
3 cited facts
On average this model trails the best reported score by 55.97 points under the disclosed evaluation harness, a wide gap that places it well behind the leaders rather than at the frontier or merely mid-pack. It holds the top score on none of its benchmarks (a leader count of 0), so there is no category leadership to claim here.
2 cited facts
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
Mistral Medium 3.5 is an AI model developed by Mistral AI, released 2026-04-28. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Mistral Medium 3.5 has recorded scores on 6 benchmarks, each shown with its evidence status.
Mistral Medium 3.5 has recorded scores on 6 benchmarks — DTBench, WeirdML, LMCA, Surface Evolver Bench, ProofBench, FrontierCode. The full table above shows each score with its evidence status.
0 of 6 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
Mistral Medium 3.5 has 6 tracked claims: 6 unverified.