Mistral Large 4 Preview
- Best result
- 98.33onOTIS Mock AIME 2024-2025unverified
- Standing
- #33 of 170 on OTIS Mock AIME 2024-2025 — top quarter of the field
- Leads
- 0 of 16 benchmarks
- Evidence
- 0 of 16 scores independently reproduced
- third-party
- unverified
Each bar is drawn on its own benchmark's scale — the ceiling that benchmark states where it has one (of 100), otherwise the best score recorded on it (rel. 2103). A benchmark that states neither draws no bar.
| Benchmark | Score | Evidence status | Released |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.33accuracy (%) | unverifiedT2 | 2026-10-06 |
| GPQA diamond | 87.59accuracy (%) | unverifiedT2 | 2026-10-06 |
| DTBench | 80.4accuracy (%) | third-partyT1 | 2026-10-06https://conceptualreasoning.ai/dtbench |
| FrontierMath-Tiers-1-3-v2-Private | 61.68accuracy (%) | unverifiedT2 | 2026-10-06 |
| APEX-Agents | 51.6accuracy (%) | unverifiedT2 | 2026-10-06 |
| LMCA | 42.11accuracy (%) | third-partyT1 | 2026-10-06https://conceptualreasoning.ai/lmca |
| MirrorCode | 40accuracy (%) | unverifiedT2 | 2026-10-06 |
| FrontierCode | 39.43accuracy (%) | third-partyT1 | 2026-10-06https://cognition.com/frontiercode |
| Chess Puzzles | 29.5accuracy (%) | unverifiedT2 | 2026-10-06 |
| Surface Evolver Bench | 28.12accuracy (%) | third-partyT1 | 2026-10-06https://yhenon.github.io/surface-evolver-llm-eval/ |
| FrontierMath-Tier-4-v2-Private | 26.83accuracy (%) | unverifiedT2 | 2026-10-06 |
| SimpleQA Verified | 20accuracy (%) | unverifiedT2 | 2026-10-06 |
| WeirdML v3 | 13.09accuracy (%) | third-partyT1 | 2026-10-06https://htihle.github.io/weirdml.html |
| ProofBench | 10accuracy (%) | third-partyT1 | 2026-10-06https://www.vals.ai/benchmarks/proof_bench |
| Mystery Game Puzzles | 8.57accuracy (%) | unverifiedT2 | 2026-10-06 |
| Furniture Assembly | 0accuracy (%) | unverifiedT2 | 2026-10-06 |
Claims drawn from cited facts, not live model generation.
This model was developed by Mistral AI, placing its origins within that organization's line of work. It arrives with a 2026 vintage, marking it as a recent entry rather than an older release.
2 cited facts
It is scored on 16 tracked benchmarks, but it holds no current top score. The record is third-party, so treat it as externally sourced rather than independently reproduced or self-reported.
3 cited facts
On its benchmarks the model trails the leading score by an average of 50.03 points under the disclosed harness — a wide gap that reads as well behind the front of the pack rather than a near-frontier miss, and it is a score gap, not a capability claim. It holds the top score on exactly 0 benchmarks, so there is no category leadership to point to — the sota leader count confirms it leads none of its evaluated tasks.
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 Large 4 Preview is an AI model developed by Mistral AI, released 2026-10-06. Its benchmark record below tags every score as independently reproduced, third-party, self-reported, or unverified.
Mistral Large 4 Preview has recorded scores on 16 benchmarks, each shown with its evidence status.
Mistral Large 4 Preview has recorded scores on 16 benchmarks — ProofBench, GPQA diamond, OTIS Mock AIME 2024-2025, DTBench, LMCA, SimpleQA Verified, and 10 more. The full table above shows each score with its evidence status.
0 of 16 tracked claims (0%) are independently reproduced; the rest are third-party, self-reported, unverified, or contradicted by an independent measurement.
Mistral Large 4 Preview has 16 tracked claims: 10 unverified.
Plain text
tensor.news. "Mistral Large 4 Preview — benchmark record". Benchmark record with evidence status for every score Updated 2026-11-18. https://tensor.news/models/mistral-large-4-preview
BibTeX
@misc{tensornews_model_mistral-large-4-preview-benchmark-record,
title = {{Mistral Large 4 Preview — benchmark record}},
author = {{tensor.news}},
year = {2026},
month = {11},
howpublished = {\url{https://tensor.news/models/mistral-large-4-preview}},
note = {Benchmark record with evidence status for every score},
urldate = {2026-11-18},
}