Mistral AI·released 2024-10-161 source
Ministral 3B benchmark scores: 4 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 4 benchmarks·best result 19.55 on DTBench (unverified)·0 of 4 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| DTBench | 19.55accuracy (%) | unverifiedT1 | 2024-10-16https://conceptualreasoning.ai/dtbench |
| MATH level 5 | 14.44accuracy (%) | unverifiedT2 | 2024-10-16 |
| LMCA | 6.45accuracy (%) | unverifiedT1 | 2024-10-16https://conceptualreasoning.ai/lmca |
| GPQA diamond | 0.34accuracy (%) | unverifiedT2 | 2024-10-16 |
Claims drawn from cited facts, not live model generation.
This model comes from Mistral AI and dates to 2024, placing it within that developer's catalogue of that year rather than an earlier or later vintage. That combination of a mid-decade release and a single named developer is the whole of what can be said about its provenance here, since no parameter count is offered to support any claim about its scale or the cost and efficiency trade-offs such a scale would imply.
4 cited facts
This model is scored on 4 tracked benchmarks but holds no current top score. Its record is unverified: outside evaluation has neither confirmed nor disputed the reported scores, so treat them as neither confirmed nor refuted.
3 cited facts
On average the model trails the benchmark leader by 80.88 points under the disclosed harness — a wide gap that reads as well behind the front of the pack, not a capability claim in itself. It holds the top score on none of its benchmarks, so there is no category leadership to report yet.
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
Ministral 3B is an AI model developed by Mistral AI, released 2024-10-16. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Ministral 3B has recorded scores on 4 benchmarks, each shown with its evidence status.
Ministral 3B has recorded scores on 4 benchmarks — GPQA diamond, MATH level 5, DTBench, LMCA. The full table above shows each score with its evidence status.
0 of 4 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
Ministral 3B has 4 tracked claims: 4 unverified.