Live
Benchmarks

PIQA

Models scored
113
Best score
0.89
Metric
Score

PIQA is an AI evaluation tracked by Epoch AI, with 113 scored model versions on record. The scores here are reported by third parties rather than produced by Epoch AI running the evaluation itself.

Scored models span November 2019 to December 2024. The best score recorded is 0.89, measured as Score. Models come from Meta AI, Mistral AI, Google DeepMind, MosaicML, Technology Innovation Institute, Baichuan, Stability AI, Alibaba and others.

Full record
Score metric
Score
Models scored
113
Best score recorded
0.89
Earliest model scored
Nov 5, 2019
Most recent model scored
Dec 26, 2024
Third-party reported
Yes
Leaderboard
01gpt-4o-mini-2024-07-18OpenAI0.89
02Phi-3.5-MoE-instructMicrosoft0.89
03gemini-1.5-flash-002Google DeepMind0.88
04Llama-3.1-405BMeta AI0.86
05falcon-180BTechnology Innovation Institute0.85
06DeepSeek-V3-BaseDeepSeek0.85
07Inflection-1Inflection AI0.84
08DeepSeek-V2DeepSeek0.84
09gemma-2-9bGoogle DeepMind0.84
10Mixtral-8x7B-v0.1Mistral AI0.84
11Mistral-Nemo-Base-2407Mistral AI0.83
12StableBeluga2Stability AI0.83
13Megatron-Turing NLG 530BMicrosoft,NVIDIA0.83
14falcon-40bTechnology Innovation Institute0.83
15Mistral-7B-v0.1Mistral AI0.83
16Llama-2-70b-hfMeta AI0.83
17LLaMA-65BMeta AI0.83
18Qwen2.5-72BAlibaba0.83
19Nemotron-4 15BNVIDIA0.82
20LLaMA-33BMeta AI0.82
21text-davinci-001OpenAI0.82
22PaLM 540BGoogle Research0.82
23Mistral-7B-Instruct-v0.2Mistral AI0.82
24Mistral-7B-Instruct-v0.1Mistral AI0.82
25Llama-2-34bMeta AI0.82
26mpt-30bMosaicML0.82
27Chinchilla (70B)DeepMind0.82
28Gopher (280B)DeepMind0.82
29gemma-7bGoogle DeepMind0.81
30Llama-3.1-8B-InstructMeta AI0.81
31Phi-3.5-mini-instructMicrosoft0.81
32Llama-2-13bMeta AI0.81
33mpt-7bMosaicML0.81
More benchmarks
SourceEpoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/benchmarks. Licensed under CC BY 4.0.
← All benchmarks