Benchgen

SciCode — Results

RankModelScore
1fugu60.1
2gemini-3-1-pro59
3fugu-ultra58.7
4kimi-k358.7
5muse-spark-open-816b-a42b58.5
6kimi-k2-652.2
7beam-1-0-501b-a23b49.2
8kimi-k2-548.7
9kimi-k2-thinking-090544.8
10nemotron-3-ultra-550b-a55b44.6
11sage-a1-544.3
12muse-glimmer43.6
13k2-horizon-375b-a23b42.7
14nemotron-3-super-120b-a12b42
15k2-horizon-mova-36b-a4b38.9
16granite-4-2-30b38.76
17granite-4-2-8b36.09
18minimax-m236
19lfm2-5-24b-a2b32.71
20k2-horizon-7b31.6
21nemotron-3-5-lightning-30b-a3b31.4
22k2-horizon-32b30.2
23k2-horizon-3-7b25.9
24granite-4-2-3b24.11

SciCode — Evaluations

1 phaseActive

No evaluations yet for this benchmark.