What does Artificial Analysis · SciCode measure?

Artificial Analysis · SciCode is a knowledge benchmark in the BenchGecko catalog. 51 AI models have been tested on it. Scores range from 14.4 to 66.9 out of 100.

Which model leads on Artificial Analysis · SciCode?

Claude Opus 5.5 from Anthropic leads Artificial Analysis · SciCode with a score of 66.9. The median score across 51 tested models is 45.5.

Is Artificial Analysis · SciCode saturated?

No · the top score is 66.9 out of 100 (67%). There is still meaningful room for improvement on Artificial Analysis · SciCode.

Does Artificial Analysis · SciCode predict performance on other benchmarks?

Yes · Artificial Analysis · SciCode scores correlate 0.94 with HLE across 5 shared models. Models that do well on Artificial Analysis · SciCode tend to do well on HLE.

How often is Artificial Analysis · SciCode data refreshed?

BenchGecko pulls updates daily. New model scores on Artificial Analysis · SciCode appear as soon as they are published by Epoch AI or the model provider.

Benchmark · KnowledgeSettled

Artificial Analysis · SciCode

Name: Artificial Analysis · SciCode Benchmark
Creator: BenchGecko
License: https://creativecommons.org/licenses/by/4.0/

Updated 2026-09-29

Models tested

Top score

66.9

Claude Opus 5.5

Median

45.5

min 14.4

Top-5 spread

σ 2.6

Competitive

The Frontier

Best score over time · one chart, every benchmark

Chart type

Frontier on Artificial Analysis · SciCode rose from 31.7 to 66.9 in 18 months · +35.2 points · latest leader Claude Opus 5.5 from Anthropic.

Pink dots = frontier records · 8 totalClick to open model page

Full rankings

51 models tested · sorted by score

#	Model	Score	Price
1	Claude Opus 5.5· Anthropic	66.9	$4.00
2	Claude Fable 5.1· Anthropic	63.1	$10.00
3	Claude Sonnet 5.5· Anthropic	61.0	$2.00
4	MiMo-V2.6-Pro· xiaomi	60.9	$0.43
5	Kimi K3· moonshotai	59.5	$2.70
6	GLM 5.3· z-ai	59.0	$1.40
7	Gemini 3.1 Pro Preview· Google DeepMind	58.7	$2.00
8	Grok 4.7· xAI	57.4	$2.00
9	Gemini 3.8 Flash· Google DeepMind	56.6	$0.75
10	GPT-6 Astra· OpenAI	56.5	$10.00
11	GPT-5.6 Terra· OpenAI	55.0	$2.00
12	GPT-6 Luna· OpenAI	54.6	$0.10
13	GPT-6.1 Sol· OpenAI	54.2	$2.00
14	Qwen3.8 2.4T A95B· Alibaba Qwen	54.1	$2.00
15	Qwen3.8 Max (0902)· Alibaba Qwen	52.1	$2.00
16	DeepSeek V4.1 Flash· DeepSeek	51.9	$0.15
17	GLM 5.3 Flash· z-ai	51.6	$0.15
18	MiMo-V2.6-Flash· xiaomi	51.3	$0.14
19	DeepSeek V4 Pro· DeepSeek	51.0	$0.21
20	Hy3· tencent	48.6	$0.13
21	Kimi K2.7 Code· moonshotai	47.8	$0.61
22	Solar Mini 4· upstage	47.6	$0.05
23	MiniMax M3· minimax	47.1	$0.30
24	Qwen3.8 27B· Alibaba Qwen	46.6	$0.42
25	Qwen3.7 Plus· Alibaba Qwen	46.1	$0.32
26	Gemma 4 31B· Google DeepMind	45.5	$0.09
27	Qwen3.5 397B A17B· Alibaba Qwen	44.8	$0.55
28	Solar Pro 4· upstage	44.6	$0.09
29	Gemini 3.5 Flash Lite· Google DeepMind	41.3	$0.30
30	Trinity Large Thinking· arcee-ai	40.6	$0.25
31	Nemotron 3 Ultra· NVIDIA	40.3	$0.50
32	Mistral Medium 3.5· Mistral AI	40.2	$1.50
33	Gemma 4 26B A4B · Google DeepMind	40.0	$0.07
34	DeepSeek V4 Pro 0813· DeepSeek	39.9	$0.66
35	Qwen3.5-122B-A10B· Alibaba Qwen	39.7	$0.26
36	gpt-oss-20b· OpenAI	38.9	$0.02
37	Mistral Small 4· Mistral AI	38.8	$0.15
38	North Mini Code (free)· Cohere	38.8	$0.00
39	Command A+· Cohere	38.5	$0.30
40	Mercury 2.5· inception	38.5	$0.04
41	Granite 4.2 8B· ibm-granite	37.8	$0.06
42	Qwen3.6 35B A3B· Alibaba Qwen	36.6	$0.15
43	LongCat 2.0· meituan	36.3	$0.30
44	Nemotron 3 Super· NVIDIA	36.2	$0.08
45	Qwen3 Coder Next· Alibaba Qwen	36.2	$0.12
46	gpt-oss-120b· OpenAI	34.0	$0.04
47	Nemotron 3.5 Lightning· NVIDIA	32.1	$0.06
48	Llama 4 Maverick· Meta	31.7	$0.19
49	Qwen3.5-9B· Alibaba Qwen	29.5	$0.10
50	Llama 4 Scout· Meta	21.3	$0.10
51	LFM2.5-2.6B (free)· liquid	14.4	$0.00