SciCode leaderboard

SciCode evaluates language models on generating code for realistic scientific research problems across 16 subfields of physics, math, chemistry, biology, and material science. Problems decompose into 338 subproblems requiring domain knowledge recall, scientific reasoning, and precise code synthesis. Based on real scripts from published research.

Models ranked by their published Scientific Code Benchmark value

Rank	Model	Provider	SciCode
1	SASakana Fugu	Sakana AI	60.1
2	SASakana Fugu-Ultra	Sakana AI	58.7
3	Qwen3.7 Max	Alibaba	53.5
4	Gemini 3.5 Flash	Google	53.1
5	Kimi K2.6	Moonshot AI	52.2
6	Qwen3.7 Plus	Alibaba	51.3
7	TMInkling-Small	Thinking Machines Lab	48.7
7	Kimi K2.5	Moonshot AI	48.7
9	Grok 4.3	xAI	47.3
10	Qwen 3.6 Max (preview)	Alibaba	47
11	Nemotron 3 Ultra	NVIDIA	44.6
12	Muse Glimmer 30B	Meta	43.6
13	Ling 3.0 Flash	InclusionAI	41.24
14	Hy3 Preview	Tencent	41.2
15	STA.X K2	SK Telecom	41
16	Ling 3.0 Flash FP8	InclusionAI	40.37
17	Granite 4.2 30B	IBM	38.76
18	Mercury 2.5	Inception	38
19	K-EXAONE 2.0	LG AI Research	37.4
20	Granite 4.2 8B	IBM	36.09
21	Nemotron 3 Nano Omni 30B A3B	NVIDIA	32
22	Nemotron 3.5 Lightning 30B A3B NVFP4	NVIDIA	31.38
23	Agents-A1-4B	InternScience	29.6
24	Ling 2.6 Flash	InclusionAI	27
25	OPMiniCPM5-2B	OpenBMB	26.3
26	Granite 4.2 3B	IBM	24.11

Evidence key: Observed

Rows are ordered by the value BenchLM published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.