# Physical Commonsense Reasoning on PIQA (PIQA Score) Source: [Physical Commonsense Reasoning on PIQA (PIQA Score)]() Collection index: [llms.txt]() Data updated: 2026-05-29T06:53:10.612Z Evaluation of model performance on the PIQA dataset for physical commonsense reasoning, reporting the primary score labeled as PIQA. 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 | Physical Commonsense Reasoning | | Dataset (including any stated split) | PIQA | | Metric | PIQA Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [PIQA Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | PIQA Score | Result date | Report paper | | --- | --- | --- | --- | | LLAMA-4-SCOUT | 81.1 | Not recorded | 691aac1f6750be83fe5ce476 | | LLAMA-4-SCOUT | 80.8 | Not recorded | 691aac1f6750be83fe5ce476 | | LLAMA-4-SCOUT | 80.7 | Not recorded | 691aac1f6750be83fe5ce476 | | QWEN-3-30B MOE | 80.5 | Not recorded | 691aac1f6750be83fe5ce476 | | QWEN-3-30B MOE | 80.4 | Not recorded | 691aac1f6750be83fe5ce476 | | DEEPSEEK-V2-LITE | 80.1 | Not recorded | 691aac1f6750be83fe5ce476 | | DEEPSEEK-V2-LITE | 79.9 | Not recorded | 691aac1f6750be83fe5ce476 | | DEEPSEEK-V2-LITE | 78.2 | Not recorded | 691aac1f6750be83fe5ce476 | | QWEN-3-30B MOE | 77.8 | Not recorded | 691aac1f6750be83fe5ce476 | | Echo-180M | 70.9 | Not recorded | 6a016293dbc836b87d823116 | | Transformer-180M | 67.1 | Not recorded | 6a016293dbc836b87d823116 | | Mamba-2-180M | 66.8 | Not recorded | 6a016293dbc836b87d823116 | | Mamba-3-MIMO-180M | 66.7 | Not recorded | 6a016293dbc836b87d823116 | | GDN-180M | 66.3 | Not recorded | 6a016293dbc836b87d823116 | | Mamba-3-SISO-180M | 66.1 | Not recorded | 6a016293dbc836b87d823116 | | Echo-50M | 59.5 | Not recorded | 6a016293dbc836b87d823116 | ### LLAMA-4-SCOUT PIQA Score: 81.1. Result ID: 6a1937cdfbc373592c130b2f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"109B","Variant":"ORIGINAL"}. This record does not assert independent reproduction. ### LLAMA-4-SCOUT PIQA Score: 80.8. Result ID: 6a1937cdfbc373592c130b30. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"109B","Variant":"FCSD"}. This record does not assert independent reproduction. ### LLAMA-4-SCOUT PIQA Score: 80.7. Result ID: 6a1937cdfbc373592c130b31. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"109B","Variant":"SFT"}. This record does not assert independent reproduction. ### QWEN-3-30B MOE PIQA Score: 80.5. Result ID: 6a1937cdfbc373592c130b2c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"30B","Variant":"ORIGINAL"}. This record does not assert independent reproduction. ### QWEN-3-30B MOE PIQA Score: 80.4. Result ID: 6a1937cdfbc373592c130b2d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"30B","Variant":"FCSD"}. This record does not assert independent reproduction. ### DEEPSEEK-V2-LITE PIQA Score: 80.1. Result ID: 6a1937cdfbc373592c130b29. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"16B","Variant":"ORIGINAL"}. This record does not assert independent reproduction. ### DEEPSEEK-V2-LITE PIQA Score: 79.9. Result ID: 6a1937cdfbc373592c130b2a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"16B","Variant":"FCSD"}. This record does not assert independent reproduction. ### DEEPSEEK-V2-LITE PIQA Score: 78.2. Result ID: 6a1937cdfbc373592c130b2b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"16B","Variant":"SFT"}. This record does not assert independent reproduction. ### QWEN-3-30B MOE PIQA Score: 77.8. Result ID: 6a1937cdfbc373592c130b2e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"PARAMS":"30B","Variant":"SFT"}. This record does not assert independent reproduction. ### Echo-180M PIQA Score: 70.9. Result ID: 6a018b6647f34ce924e3238a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"180M","Tokens":"10B","Zero-shot":"true"}. This record does not assert independent reproduction. ### Transformer-180M PIQA Score: 67.1. Result ID: 6a018b6647f34ce924e3238c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"180M","Tokens":"100B","Zero-shot":"true"}. This record does not assert independent reproduction. ### Mamba-2-180M PIQA Score: 66.8. Result ID: 6a018b6647f34ce924e3238e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"180M","Tokens":"100B","Zero-shot":"true"}. This record does not assert independent reproduction. ### Mamba-3-MIMO-180M PIQA Score: 66.7. Result ID: 6a018b6647f34ce924e32390. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"180M","Tokens":"100B","Zero-shot":"true"}. This record does not assert independent reproduction. ### GDN-180M PIQA Score: 66.3. Result ID: 6a018b6647f34ce924e3238d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"180M","Tokens":"100B","Zero-shot":"true"}. This record does not assert independent reproduction. ### Mamba-3-SISO-180M PIQA Score: 66.1. Result ID: 6a018b6647f34ce924e3238f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"180M","Tokens":"100B","Zero-shot":"true"}. This record does not assert independent reproduction. ### Echo-50M PIQA Score: 59.5. Result ID: 6a018b6647f34ce924e3238b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Params":"50M","Tokens":"3B","Zero-shot":"true"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.