Open LLM Leaderboard v1 - WinoGrande: leaderboard
Original Hugging Face Open LLM Leaderboard WinoGrande pronoun-resolution questions, scored by multiple-choice accuracy with five demonstrations. Percentage, higher is better; two-choice chance is 50%. Frozen v1 archive; public questions may overlap model training.
Metric: Accuracy (%) (5-shot). Source: huggingface.co. Saturation forecast: Estimated already saturated. 7252 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | LogoS-7Bx2-MoE-13B-v0.2 | 88.4 | #873 |
| 2 | Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B | 88.24 | #848 |
| 3 | DARE_TIES_13B | 88.08 | #856 |
| 4 | MoE_13B_DPO | 88 | #857 |
| 5 | Rhea-72B-v0.5 | 87.85 | #555 |
| 6 | free-evo-qwen72B-v0.8-re | 87.77 | #442 |
| 7 | MixTAO-7Bx2-MoE-v8.1 | 87.37 | #815 |
| 8 | WestLake-7B-v2 | 86.98 | #877 |
| 9 | WestLake-7B-v2-laser | 86.74 | #886 |
| 10 | llamaRAGdrama | 86.66 | #881 |
| 11 | multimaster-7B-v6 | 86.42 | #896 |
| 12 | luxia-21.4B-alignment-v1.2 | 86.27 | #963 |
| 13 | mera-mix-4x7B | 85.64 | #930 |
| 14 | internlm2-20B | 85.56 | #937 |
| 15 | Le_Triomphant-ECE-TW3 | 85.56 | #564 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #6 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-llm-leaderboard-v1-winogrande · How It Works · Data refreshed daily, snapshot 2026-11-07.