Skip to content

Benchmark

SkillsBench

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

Models and systems measured
27
Scores recorded
32
Sources
3
IQ weight
1.99

Score distribution

54.4%
Mean
55.3%
Median
10.7%
Std dev
44.3%
10th percentile
68.7%
90th percentile
28.7%
Min
28.7%73.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.99 for chat models, measured over 26 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
1GPT-5.6 SolLing-3.1-flash official chart · SkillsBench · version, task count, harness, reasoning setting and run count not stated· Reported by Ant inclusionAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.73.5%
2DeepSeek V4.1 FlashLing-3.1-flash official chart · SkillsBench · version, task count, harness, reasoning setting and run count not stated· Reported by Ant inclusionAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.71.3%
3Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim70.2%
4Ling 3.1 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim68.7%
5Ling 3.1 FlashLing-3.1-flash official chart · SkillsBench · version, task count, harness, reasoning setting and run count not stated· Reported by Ant inclusionAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.68.7%
6GPT 5.5Dots Studio official appendix · unstarred comparison cell · version, task count, harness and reasoning setting not independently specified· Reported by Dots Studio / Xiaohongshu· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.66.5%
7Claude Opus 5Ling-3.1-flash official chart · SkillsBench · version, task count, harness, reasoning setting and run count not stated· Reported by Ant inclusionAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.63.7%
8Kimi K3Ling-3.1-flash official chart · SkillsBench · version, task count, harness, reasoning setting and run count not stated· Reported by Ant inclusionAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.63.6%
9LongCat 2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim63.3%
10Hy4 preview· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
62.9%
11GLM 5.3 FlashLing-3.1-flash official chart · SkillsBench · version, task count, harness, reasoning setting and run count not stated· Reported by Ant inclusionAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.61.3%
12GLM 5.3Ling-3.1-flash official chart · SkillsBench · version, task count, harness, reasoning setting and run count not stated· Reported by Ant inclusionAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.59.4%
13Qwen3.7 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim59.2%
14GLM 5.1· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim58.4%
15Hy3 (Deep Infra)· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
55.3%
16Hy3 (Deep Infra)Dots Studio official appendix · unstarred comparison cell · version, task count, harness and reasoning setting not independently specified· Reported by Dots Studio / Xiaohongshu· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.55.3%
17Qwen3.7 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim54.9%
18Kimi K2.6· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim54%
19MiniMax M3· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim53%
20DeepSeek V4 Pro· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim50.1%
21Qwen3.6 27B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
48.2%
22Qwen3.6 Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim45.7%
23Ling 3.0 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim44.8%
24DeepSeek V4 Flash· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim44.7%
25DeepSeek V4 FlashDots Studio official appendix · unstarred comparison cell · version, task count, harness and reasoning setting not independently specified· Reported by Dots Studio / Xiaohongshu· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.44.7%
26Muse Glimmer 30B· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
44.3%
27Muse Glimmer 30Bwith skills · high reasoning· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.44.3%
28openPangu 2.0 FlashopenPangu official table · Thinking · Avg@5 · version, task subset, harness and detailed thinking budget not stated· Reported by Huawei openPangu· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.42.6%
29openPangu 2.0 FlashopenPangu official table · Non-Thinking · Avg@5 · version, task subset, harness and detailed thinking budget not stated· Reported by Huawei openPangu· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.40%
30MiniMax M2.7· Hugging Face Official Benchmarks data· Reported by the model's developer. The source relaying it did not verify it. claim34.9%
31Qwen3.6 35B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim28.7%

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
DeepSeek V4 Pro Preview (SkillsBench table label)Dots Studio official appendix · unstarred comparison cell · version, task count, harness and reasoning setting not independently specified· Reported by Dots Studio / Xiaohongshu· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.50.1%

Data credits: 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. · Hugging Face official benchmark leaderboards; licence varies by benchmark · Data by LLM Stats