PIQA
Rank #47 of 51 · 44 models scored · top score 79.1
unknown
Strongest on discrimination, weakest on contamination resistance. Caveat: scores come from mixed harnesses, so a head-to-head is not apples-to-apples.
Weighted composite. Each signal is computed from the source-backed leaderboard below.
Does it still separate models?
How far from ceiling / clustered at the top?
Public vs held-out; training-leak risk.
Apples-to-apples, or mixed / vendor-optimized?
How old is the benchmark?
| # | Model | Score | Evidence | Measured |
|---|---|---|---|---|
| 1 | PowerMoE-3b | 79.1 | unverified | — |
| 2 | GPT-4o mini | 77.4 | self-reported· optimized | 2024-07-18 |
| 3 | Gemini 1.5 Flash (Sep 2024) | 75 | self-reported· optimized | 2024-05-10 |
| 4 | Llama 3.1-405B | 71.8 | self-reported· optimized | 2024-07-23 |
| 5 | Falcon-180B | 69.8 | unverified· optimized | 2023-09-06 |
| 6 | DeepSeek-V3 | 69.4 | self-reported· optimized | 2024-12-24 |
| 7 | DeepSeek-V2 (MoE-236B, May 2024) | 67.8 | self-reported· optimized | 2024-05-07 |
| 8 | Gemma 2 9B | 67.4 | self-reported· optimized | 2024-06-24 |
| 9 | Mixtral 8x7B | 67.2 | unverified· optimized | 2023-12-11 |
| 10 | Mistral NeMo | 67 | self-reported· optimized | 2024-07-18 |
| 11 | Stable Beluga 2 | 66.6 | self-reported· optimized | 2023-07-20 |
| 12 | Mistral 7B v0.1 | 66 | self-reported· optimized | 2023-10-10 |
| 13 | Falcon-40B | 66 | unverified· optimized | 2023-03-15 |
| 14 | LLaMA-65B | 65.6 | self-reported· optimized | 2023-02-24 |
| 15 | Llama 2-70B | 65.6 | self-reported· optimized | 2023-07-18 |
| 16 | Qwen2.5-72B | 65.2 | self-reported· optimized | 2024-09-19 |
| 17 | Nemotron-4 15B | 64.8 | self-reported· optimized | 2024-02-27 |
| 18 | LLaMA-33B | 64.6 | self-reported· optimized | 2023-02-27 |
| 19 | Llama 2-34B | 63.8 | unverified· optimized | 2023-07-18 |
| 20 | MPT-30B | 63.8 | unverified· optimized | 2023-06-22 |
| 21 | Gemma 7B | 62.4 | unverified· optimized | 2024-02-21 |
| 22 | Llama 3.1-8B | 62.4 | self-reported· optimized | 2024-07-23 |
| 23 | Llama 2-13B | 61.6 | unverified· optimized | 2023-07-18 |
| 24 | MPT-7B | 61.2 | self-reported· optimized | 2023-05-05 |
| 25 | internlm-20b | 60.6 | self-reported· optimized | 2023-09-18 |
| 26 | Falcon-7B | 60.6 | self-reported· optimized | 2023-04-24 |
| 27 | LLaMA-13B | 60.2 | self-reported· optimized | 2023-02-27 |
| 28 | Qwen-14B | 59.8 | self-reported· optimized | 2023-09-24 |
| 29 | LLaMA-7B | 59.6 | self-reported· optimized | 2023-02-24 |
| 30 | Llama 2-7B | 57.6 | unverified· optimized | 2023-07-18 |
| 31 | Baichuan2-13B | 56.2 | self-reported· optimized | 2023-09-06 |
| 32 | Qwen-7B | 55.8 | self-reported· optimized | 2023-09-28 |
| 33 | internlm-7b | 55.8 | self-reported· optimized | 2023-07-05 |
| 34 | vicuna-13b-v1.1 | 54.8 | self-reported· optimized | 2023-04-12 |
| 35 | Gemma 2B | 54.6 | unverified· optimized | 2024-02-21 |
| 36 | RedPajama-INCITE-7B-Base | 53.8 | self-reported· optimized | 2023-05-04 |
| 37 | Baichuan1-7B | 52.4 | self-reported· optimized | 2023-06-01 |
| 38 | open_llama_7b | 52 | self-reported· optimized | 2023-06-07 |
| 39 | XGen-7B | 51 | self-reported· optimized | 2023-09-07 |
| 40 | Dolly 2.0-12b | 50.8 | self-reported· optimized | 2023-04-12 |
| 41 | Cerebras-GPT-13B | 47 | self-reported· optimized | 2023-04-06 |
| 42 | Qwen-1_8B | 46.6 | self-reported· optimized | 2023-11-30 |
| 43 | chatglm2-6b | 39.2 | self-reported· optimized | 2023-06-24 |
| 44 | stablelm-tuned-alpha-7b | 31.6 | self-reported· optimized | 2023-04-19 |