Mistral AI·released 2023-09-271 source
Mistral 7B v0.1 benchmark scores: 10 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 10 benchmarks·best result 75.2 on TriviaQA (unverified)·0 of 10 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| TriviaQA | 75.2accuracy (%) | unverified· optimizedT1 | 2023-09-27Gemma: Open Models Based on Gemini Research and Technology |
| HellaSwag | 74.67accuracy (%) | unverified· optimizedT1 | 2023-09-27Mixtral of Experts |
| OpenBookQA | 73.07accuracy (%) | self-reported· optimizedT1 | 2023-09-27Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone |
| ARC AI2 | 71.47accuracy (%) | self-reported· optimizedT1 | 2023-09-27Nemotron-4 15B Technical Report |
| PIQA | 66accuracy (%) | self-reported· optimizedT1 | 2023-09-27Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone |
| GSM8K | 54.4accuracy (%) | unverified· optimizedT1 | 2023-09-27Stanford HELM |
| Winogrande | 50.6accuracy (%) | unverified· optimizedT1 | 2023-09-27Training Compute-Optimal Large Language Models |
| MMLU | 50accuracy (%) | unverified· optimizedT1 | 2023-09-27Mixtral of Experts |
| BBH | 41.47accuracy (%) | unverified· optimizedT1 | 2023-09-27Gemma: Open Models Based on Gemini Research and Technology |
| ANLI | 20.65accuracy (%) | self-reported· optimizedT1 | 2023-09-27Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone |
Claims drawn from cited facts, not live model generation.
Developed by Mistral AI and released in 2023, this model has 7.0 billion parameters, making it a compact-scale model. That compact scale implies a trade-off of better efficiency and lower cost, rather than the greater headroom and expense of larger models.
4 cited facts
This model is scored on 10 tracked benchmarks, but it holds no current top score. Its record is self-reported, meaning the numbers are vendor-claimed and not yet independently confirmed, so they should be read as claims rather than verified results.
3 cited facts
Across its benchmarks the model trails the best reported score by an average of 23.79 points — a wide gap under the disclosed harness that reads as well behind the leaders rather than at the frontier. It also holds the top score on 0 of those benchmarks, so by this reading there is no category leadership to point to yet.
2 cited facts
2 facts cross-checked across data sources: 2 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: curated_param_counts, epoch_benchmarks
Mistral 7B v0.1 is an AI model developed by Mistral AI, released 2023-09-27. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Mistral 7B v0.1 has recorded scores on 10 benchmarks, each shown with its evidence status.
Mistral 7B v0.1 has recorded scores on 10 benchmarks — PIQA, OpenBookQA, Winogrande, TriviaQA, MMLU, ARC AI2, and 4 more. The full table above shows each score with its evidence status.
0 of 10 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
Mistral 7B v0.1 has 10 tracked claims: 4 self-reported, 6 unverified.