Benchgen

Winogrande — Results

RankModelScore
1gpt-4-06130.875
2mimo-v2-5-pro0.856
3command-r-plus0.854
4qwen2-72b-instruct0.851
5llama-3-1-nemotron-70b-instruct0.845
6gemma-2-27b0.837
7hermes-3-70b0.832
8qwen2-5-32b-instruct0.82
9phi-3-5-moe-instruct0.813
10qwen2-5-coder-32b-instruct0.808
11gemma-2-9b0.806
12mistral-nemo-instruct0.768
13ministral-8b-instruct0.753
14granite-3-3-8b-base0.744
15qwen2-5-coder-7b-instruct0.729
16gemma-3n-e4b-litert-preview0.717
17gemma-3n-e4b0.717
18phi-3-5-mini-instruct0.685
19phi-4-mini0.67
20gemma-3n-e2b-litert-preview0.668
21gemma-3n-e2b0.668
22ernie-4-50.513

Winogrande — Evaluations

1 phaseActive

No evaluations yet for this benchmark.