Winogrande
Rank #48 of 51 · 64 models scored · top score 78.4
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 | Llama 3.1-405B | 78.4 | unverified· optimized | 2024-07-23 |
| 2 | Claude 3 Opus | 77 | unverified· optimized | 2024-03-04 |
| 3 | GPT-4 (Mar 2023) | 75 | self-reported· optimized | 2023-03-15 |
| 4 | Falcon-180B | 74.2 | unverified· optimized | 2023-09-06 |
| 5 | DeepSeek-V2 (MoE-236B, May 2024) | 72.6 | self-reported· optimized | 2024-05-07 |
| 6 | DeepSeek-V3 | 70.4 | self-reported· optimized | 2024-12-24 |
| 7 | Llama 3-70B | 67 | unverified· optimized | 2024-04-18 |
| 8 | PaLM 2-L | 66 | self-reported· optimized | 2023-05-17 |
| 9 | PowerMoE-3b | 65 | unverified | — |
| 10 | Qwen2.5-72B | 64.6 | self-reported· optimized | 2024-09-19 |
| 11 | GPT-3.5 Turbo (Jun 2023) | 63.2 | unverified· optimized | 2023-06-13 |
| 12 | phi-3-medium 14B | 63 | self-reported· optimized | 2024-04-23 |
| 13 | phi-3-small 7.4B | 63 | self-reported· optimized | 2024-04-23 |
| 14 | Qwen2.5-Coder-32B | 61.6 | self-reported· optimized | 2024-09-18 |
| 15 | Llama 2-70B | 60.4 | unverified· optimized | 2023-07-18 |
| 16 | PaLM 2-M | 58.4 | self-reported· optimized | 2023-05-17 |
| 17 | Gemma 7B | 58 | unverified· optimized | 2024-02-21 |
| 18 | Falcon 2 11B | 56.6 | self-reported· optimized | 2024-05-09 |
| 19 | Nemotron-4 15B | 56 | self-reported· optimized | 2024-02-27 |
| 20 | PaLM 2-S | 55.8 | self-reported· optimized | 2023-05-17 |
| 21 | Mixtral 8x7B | 54.4 | unverified· optimized | 2023-12-11 |
| 22 | LLaMA-65B | 54 | unverified· optimized | 2023-02-24 |
| 23 | Falcon-40B | 53.8 | unverified· optimized | 2023-03-15 |
| 24 | Qwen2.5-Coder-14B | 53.6 | self-reported· optimized | 2024-09-18 |
| 25 | Llama 2-34B | 53.4 | unverified· optimized | 2023-07-18 |
| 26 | LLaMA-33B | 52 | unverified· optimized | 2023-02-27 |
| 27 | Llama 3-8B | 51.4 | unverified· optimized | 2024-04-18 |
| 28 | Mistral 7B v0.1 | 50.6 | unverified· optimized | 2023-10-10 |
| 29 | Claude 3 Sonnet | 50.2 | unverified· optimized | 2024-03-04 |
| 30 | Claude 3 Haiku | 48.4 | unverified· optimized | 2024-03-04 |
| 31 | Phi-1.5 | 46.8 | self-reported· optimized | 2023-09-11 |
| 32 | LLaMA-13B | 46 | unverified· optimized | 2023-02-27 |
| 33 | Yi-9B | 46 | unverified· optimized | 2024-03-01 |
| 34 | Qwen2.5-Coder (7B) | 45.8 | self-reported· optimized | 2024-09-18 |
| 35 | DeepSeek-Coder-V2-Lite-Base | 45.8 | self-reported· optimized | 2024-06-13 |
| 36 | Llama 2-13B | 45.6 | unverified· optimized | 2023-07-18 |
| 37 | Yi 6B | 42.6 | unverified· optimized | 2023-11-02 |
| 38 | MPT-30B | 42 | unverified· optimized | 2023-06-22 |
| 39 | phi-3-mini 3.8B | 41.6 | self-reported· optimized | 2024-04-23 |
| 40 | vicuna-13b-v1.1 | 41.6 | self-reported· optimized | 2023-04-12 |
| 41 | LLaMA-7B | 40.2 | unverified· optimized | 2023-02-24 |
| 42 | Llama 2-7B | 38.4 | unverified· optimized | 2023-07-18 |
| 43 | GPT-3.5 Turbo (Nov 2023) | 37.6 | self-reported· optimized | 2023-06-13 |
| 44 | MPT-7B | 37.2 | unverified· optimized | 2023-05-05 |
| 45 | Qwen2.5-Coder-3B | 34.8 | self-reported· optimized | 2024-09-18 |
| 46 | Falcon-7B | 34.4 | unverified· optimized | 2023-04-24 |
| 47 | open_llama_7b | 34 | self-reported· optimized | 2023-06-07 |
| 48 | INTELLECT-1 | 31.64 | self-reported· optimized | 2024-11-29 |
| 49 | Gemma 2B | 30.8 | unverified· optimized | 2024-02-21 |
| 50 | XGen-7B | 29.8 | self-reported· optimized | 2023-09-07 |
| 51 | StarCoder 2 15B | 28.6 | self-reported· optimized | 2024-02-29 |
| 52 | RedPajama-INCITE-7B-Base | 27.6 | self-reported· optimized | 2023-05-04 |
| 53 | DeepSeek Coder 33B | 24 | self-reported· optimized | 2024-01-25 |
| 54 | Dolly 2.0-12b | 23.6 | self-reported· optimized | 2023-04-12 |
| 55 | Cerebras-GPT-13B | 21.6 | self-reported· optimized | 2023-04-06 |
| 56 | Qwen2.5-Coder (1.5B) | 21.4 | self-reported· optimized | 2024-09-18 |
| 57 | CodeQwen1.5-7B | 19.6 | self-reported· optimized | 2024-04-15 |
| 58 | DeepSeek Coder 6.7B | 15.2 | self-reported· optimized | 2024-01-25 |
| 59 | StarCoder 2 3B | 14.2 | self-reported· optimized | 2024-02-29 |
| 60 | StarCoder 2 7B | 14.2 | self-reported· optimized | 2024-02-29 |
| 61 | Qwen2.5-Coder-0.5B | 9.6 | self-reported· optimized | 2024-09-18 |
| 62 | Phi-2 | 9.4 | self-reported· optimized | 2023-12-12 |
| 63 | DeepSeek Coder 1.3B | 6.6 | self-reported· optimized | 2024-01-25 |
| 64 | stablelm-tuned-alpha-7b | 3 | self-reported· optimized | 2023-04-19 |