SciCode (subtask)
SciCode (subtask) - scientific code generation benchmark, subtask-level scoring.
Category
STEM & Reasoning
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for SciCode (subtask)
| Model | Score | Date | Verified |
|---|---|---|---|
| NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16 |
30.58%
|
04.06.2026 | Verified |
| MiniMax-M2.7 |
38.30
|
04.06.2026 | Unverified |
| GLM-5.1 |
45.63%
|
04.06.2026 | Unverified |
| Kimi K2.6 |
66.50%
|
04.06.2026 | Unverified |
| Qwen3.5-397B-A17B |
47.09%
|
04.06.2026 | Unverified |
| DeepSeek-V4-Pro |
59.22%
|
04.06.2026 | Unverified |
| DeepSeek-V4-Flash |
48.06%
|
04.06.2026 | Unverified |
| Muse-Glimmer-30B |
25.73%
|
01.08.2026 | Verified |
| Gemma4-31B-it |
24.76%
|
01.08.2026 | Unverified |
| Qwen3.6-27B |
7.28%
|
01.08.2026 | Unverified |
| Kimi K3 |
99.03%
|
18.08.2026 | Verified |
| GPT-5.4 |
88.83%
|
23.08.2026 | Verified |
| Claude Opus 4.6 |
66.02%
|
23.08.2026 | Verified |
| Gemini 3.1 Pro |
100.00%
|
23.08.2026 | Verified |
| Kimi K2.5 |
50.49%
|
23.08.2026 | Verified |