# Broad Knowledge Question Answering on MMLU (Pass@1) Source: [Broad Knowledge Question Answering on MMLU (Pass@1)]() Collection index: [llms.txt]() Data updated: 2026-04-14T08:30:54.521Z Evaluation of model performance on the MMLU dataset for broad knowledge question answering, measured using Pass@1 accuracy. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Broad Knowledge Question Answering | | Dataset (including any stated split) | MMLU | | Metric | Pass@1 Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Pass@1 Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Pass@1 Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Qwen2.5-7B | 61.3 | Not recorded | 68fa7bc11e209131cea8a35b | | Qwen2.5-7B + PPO w/ VERL. | 61 | Not recorded | 68fa7bc11e209131cea8a35b | | Qwen2.5-7B + GRPO w/ VERL. | 60.9 | Not recorded | 68fa7bc11e209131cea8a35b | | Qwen2.5-7B + PPO | 60.9 | Not recorded | 68fa7bc11e209131cea8a35b | | Qwen2.5-7B + GRPO | 60.6 | Not recorded | 68fa7bc11e209131cea8a35b | | Mathstral-7B-v0.1 + GRPO w/ VERL. | 58.8 | Not recorded | 68fa7bc11e209131cea8a35b | | Mathstral-7B-v0.1 + PPO w/ VERL. | 56.9 | Not recorded | 68fa7bc11e209131cea8a35b | | Mathstral-7B-v0.1 + GRPO | 56.7 | Not recorded | 68fa7bc11e209131cea8a35b | | Mathstral-7B-v0.1 + PPO | 56.5 | Not recorded | 68fa7bc11e209131cea8a35b | | Mathstral-7B-v0.1 | 53.9 | Not recorded | 68fa7bc11e209131cea8a35b | ### Qwen2.5-7B Pass@1 Accuracy: 61.3. Result ID: 69ddfb2147f34ce924c562c8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### Qwen2.5-7B + PPO w/ VERL. Pass@1 Accuracy: 61. Result ID: 69ddfb2147f34ce924c562cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Training":"PPO w/ VERL."}. This record does not assert independent reproduction. ### Qwen2.5-7B + GRPO w/ VERL. Pass@1 Accuracy: 60.9. Result ID: 69ddfb2147f34ce924c562ca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Training":"GRPO w/ VERL."}. This record does not assert independent reproduction. ### Qwen2.5-7B + PPO Pass@1 Accuracy: 60.9. Result ID: 69ddfb2147f34ce924c562cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Training":"PPO"}. This record does not assert independent reproduction. ### Qwen2.5-7B + GRPO Pass@1 Accuracy: 60.6. Result ID: 69ddfb2147f34ce924c562c9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Training":"GRPO"}. This record does not assert independent reproduction. ### Mathstral-7B-v0.1 + GRPO w/ VERL. Pass@1 Accuracy: 58.8. Result ID: 69ddfb2147f34ce924c562c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Mathstral-7B-v0.1","Training":"GRPO w/ VERL."}. This record does not assert independent reproduction. ### Mathstral-7B-v0.1 + PPO w/ VERL. Pass@1 Accuracy: 56.9. Result ID: 69ddfb2147f34ce924c562c7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Mathstral-7B-v0.1","Training":"PPO w/ VERL."}. This record does not assert independent reproduction. ### Mathstral-7B-v0.1 + GRPO Pass@1 Accuracy: 56.7. Result ID: 69ddfb2147f34ce924c562c4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Mathstral-7B-v0.1","Training":"GRPO"}. This record does not assert independent reproduction. ### Mathstral-7B-v0.1 + PPO Pass@1 Accuracy: 56.5. Result ID: 69ddfb2147f34ce924c562c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Mathstral-7B-v0.1","Training":"PPO"}. This record does not assert independent reproduction. ### Mathstral-7B-v0.1 Pass@1 Accuracy: 53.9. Result ID: 69ddfb2147f34ce924c562c3. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Mathstral-7B-v0.1"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.