Skip to content

Benchmark

SciCode

Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.

Models measured
154
Scores recorded
265
Sources
3
IQ weight
1.71

Score distribution

47.7%
Mean
50.1%
Median
10.7%
Std dev
35.4%
10th percentile
58.8%
90th percentile
1.3%
Min
1.3%91.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.71 for chat models, measured over 154 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Mistral Large 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
91.8%
2Claude Fable 5.2max· Epoch AI Benchmarking Hub data67.7%
3Claude Opus 5.5max· Epoch AI Benchmarking Hub data66.9%
4Claude Fable 5.2xhigh· Epoch AI Benchmarking Hub data65.6%
5Claude Opus 5.5xhigh· Epoch AI Benchmarking Hub data65%
6Gemini 4 Carbonhigh· Epoch AI Benchmarking Hub data64.2%
7Claude Fable 5.1max· Epoch AI Benchmarking Hub data63.1%
8Gemini 4 Argonhigh· Epoch AI Benchmarking Hub data61.8%
9Claude Fable 5max· Epoch AI Benchmarking Hub data61%
10Claude Sonnet 5.5max· Epoch AI Benchmarking Hub data61%
11Claude Fable 5.1xhigh· Epoch AI Benchmarking Hub data60.9%
12MiMo V2.6 Pro· Epoch AI Benchmarking Hub data60.9%
13Kimi K3.1 Pro· Epoch AI Benchmarking Hub data60.8%
14Claude Opus 5.5high· Epoch AI Benchmarking Hub data60.4%
15Muse Wonder 1.0max· Epoch AI Benchmarking Hub data60.3%
16Gemini 3.7 Flashmedium· Epoch AI Benchmarking Hub data59.8%
17Doubao Seed 2.1 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim59.8%
18Claude Fable 5.2medium· Epoch AI Benchmarking Hub data59.7%
19Muse Spark 1.3xhigh· Epoch AI Benchmarking Hub data59.7%
20Claude Fable 5.2high· Epoch AI Benchmarking Hub data59.5%
21Kimi K3max· Epoch AI Benchmarking Hub data59.5%
22Claude Opus 5.5medium· Epoch AI Benchmarking Hub data59.3%
23GLM 5.3max· Epoch AI Benchmarking Hub data59%
24Gemini 3.1 Pro Previewsetting not named by the source· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.59%
25Gemini 3.1 Pro Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
59%
26Gemini 4 Argonmedium· Epoch AI Benchmarking Hub data58.9%
27Step 5 Preview· Epoch AI Benchmarking Hub data58.9%
28Muse Spark 1.1· Epoch AI Benchmarking Hub data58.8%
29Muse Spark 1.3max· Epoch AI Benchmarking Hub data58.8%
30Fugu UltraFugu official headline evaluation· Reported by Sakana AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.58.7%
31Kimi K3max effort · temp 1.0 · top_p .95· Reported by Moonshot AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.58.7%
32Claude Fable 5.1high· Epoch AI Benchmarking Hub data58.7%
33Kimi K3· Epoch AI Benchmarking Hub data58.7%
34Claude Opus 5.5low· Epoch AI Benchmarking Hub data58.6%
35Muse Spark Open· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim58.5%
36Muse Spark Open 816B A42Bxhigh· Epoch AI Benchmarking Hub data58.4%
37Doubao Seed 2.1 Turbo· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim57.8%
38Claude Fable 5.2low· Epoch AI Benchmarking Hub data57.8%
39GPT-5.6 Solhigh· Epoch AI Benchmarking Hub data57.8%
40Grok 4.7high· Epoch AI Benchmarking Hub data57.8%
41GPT-6 Solmax· Epoch AI Benchmarking Hub data57.6%
42GPT-5.6 Solmedium· Epoch AI Benchmarking Hub data57.4%
43Grok 4.7xhigh· Epoch AI Benchmarking Hub data57.4%
44Muse Spark 1.2xhigh· Epoch AI Benchmarking Hub data57.4%
45Claude Sonnet 5.5xhigh· Epoch AI Benchmarking Hub data57.3%
46Gemini 3.7 Flashhigh· Epoch AI Benchmarking Hub data57.2%
47GPT-5.6 Solmax· Epoch AI Benchmarking Hub data57.1%
48GPT-5.6 Solxhigh· Epoch AI Benchmarking Hub data57.1%
49Claude Fable 5.1low· Epoch AI Benchmarking Hub data56.7%
50Gemini 3.8 Flashhigh· Epoch AI Benchmarking Hub data56.6%

Showing 1–50 of 265 results.

Data credits: Data by LLM Stats · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.