Benchgen

MBPP — Results

RankModelScore
1llama-3-3-nemotron-super-49b-v10.913
2qwen2-5-coder-32b-instruct0.902
3qwen2-5-72b-instruct0.882
4qwen2-5-32b-instruct0.84
5qwen2-5-vl-32b-instruct0.84
6qwen2-5-coder-7b-instruct0.835
7qwen2-5-14b-instruct0.82
8qwen3-235b-a22b0.814
9phi-3-5-moe-instruct0.808
10qwen2-72b-instruct0.802
11qwen2-5-7b-instruct0.792
12llama-4-maverick0.776
13mistral-small-3-1-24b-instruct0.747
14qwen2-5-omni-7b0.732
15mistral-small-3-24b-base0.696
16phi-3-5-mini-instruct0.696
17llama-4-scout0.678
18qwen2-7b-instruct0.672

MBPP — Leaderboard

1 phaseActive

MBPP Leaderboard

No self-reported entries.