Skip to content

Benchmark

Vision2Web

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

Models measured
9
Scores recorded
14
Sources
2
IQ weight
1.89

Score distribution

62.5%
Mean
64%
Median
11.9%
Std dev
42.1%
10th percentile
77.8%
90th percentile
31%
Min
31%78.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.89 for chat models, measured over 9 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1LongCat 2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim78.6%
2GLM 5.3 Flashofficial GLM-5.3 Flash release table· Reported by Z.ai· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.77.8%
3Qwen3.8 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69%
4Qwen3.8 MaxQwen3.8-Max release vision/video table · frontend/webpage/website average · Claude Code · gpt-5.4-2026-03-05 judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.69%
5Qwen 3.8 Flash Next· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64%
6Qwen 3.8 Flash NextQwen3.8-Flash-Next release/model card · frontend/webpage/website average · Claude Code · gpt-5.4-2026-03-05 judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64%
7Qwen3.8 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64%
8Qwen3.8 FlashFrontend/Webpage/Website 平均 · Claude Code · gpt-5.4-2026-03-05 judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.64%
9Qwen 3.8 27B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim62.9%
10Qwen 3.8 27BFrontend/Webpage/Website 平均 · Claude Code · gpt-5.4-2026-03-05 judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.9%
11Qwen 3.8 27BQwen3.8-27B model card · frontend/webpage/website average · Claude Code · gpt-5.4-2026-03-05 judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.9%
12Qwen3.8 Omni FlashFrontend/Webpage/Website 平均 · Claude Code · gpt-5.4-2026-03-05 judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.62.9%
13Qwen3.7 PlusFrontend/Webpage/Website 平均 · Claude Code · gpt-5.4-2026-03-05 judge· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.42.1%
14GLM 5V Turbo· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim31%

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. · Data by LLM Stats