SciCode is an AI benchmark used to evaluate model capabilities. Review its overview, metrics, official resources, and model leaderboard results on DataLearnerAI.
Browse the latest scores, model modes, release dates, and parameter sizes for SciCode.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
— |  Claude Fable 5 Thinking Level · Max | 60.19 | 2026-06-09 | Unknown | Closed |
— |  Seed2.1 Pro Thinking Enabled | 59.80 | 2026-06-23 | Unknown | Closed |
— |  Kimi K3 Thinking Level · MaxTools | 58.70 | 2026-07-16 | 2800B | Conditional |
— |  GPT-5.6 Sol Thinking Level · High | 56.94 | 2026-06-26 | Unknown | Closed |
— | GPT-5.6 Sol Thinking Level · Medium | 56.48 | 2026-06-26 | Unknown | Closed |
— | GPT-5.6 Sol Thinking Level · Max | 56.13 | 2026-06-26 | Unknown | Closed |
— | GPT-5.6 Terra Thinking Level · Max | 53.94 | 2026-06-26 | Unknown | Closed |
— |  Qwen3.7 Max Thinking Level · Max | 53.50 | 2026-05-20 | Unknown | Closed |
— | GPT-5.6 Terra Thinking Level · Extra High | 51.62 | 2026-06-26 | Unknown | Closed |
— |  MiniMax M3 Thinking Enabled | 45.37 | 2026-06-01 | 428B | Non-Commercial |
— |  Nemotron 3 Ultra Thinking Enabled | 44.60 | 2026-06-04 | 550B | Free Commercial |
— |  Muse Glimmer-30B Thinking Level · High | 43.60 | 2026-08-10 | 29.6B | Free Commercial |
— |  Mistral Medium 3.5 Thinking Enabled | 39.58 | 2026-05-01 | Unknown | Closed |
— | GPT OSS 120B Thinking Level · High | 38.89 | 2025-08-06 | 11.7B | Free Commercial |
— | Spark-X2.5-4B Thinking Enabled | 34.70 | 2026-09-01 | 4.1B | Free Commercial |
— |  MiniCPM5-2B unknown | 26.30 | 2026-09-07 | 2.5B | Free Commercial |
