Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Web Agent Tasks on Mind2Web
Loading...
60.3
Success Rate
Gemini-3-Pro
25.356
34.428
43.5
52.572
Jun 16, 2026
Success Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Success Rate
Gemini-3-Pro
2026.06
60.3
Claude-Opus-4.5
2026.06
57.9
GPT-5.1
2026.06
55.1
Qwen3-Coder-480B-A35B-Instruct
2026.06
54
GLM-4.7
2026.06
53.7
Kimi-K2-Instruct-0905
2026.06
53.4
Qwen3-235B-A22B-Instruct-2507
2026.06
49
DeepSeek-V3.2
2026.06
47.2
Qwen2.5-Coder-14B-Instruct
2026.06
42.7
LoopCoder-v2
R (number of refinemen...
2026.06
41.4
Qwen2.5-Coder-7B-Instruct
2026.06
38.4
Seed-Coder-8B-Instruct
2026.06
38.2
LoopCoder-v2
R (number of refinemen...
2026.06
35.3
LoopCoder-v2
R (number of refinemen...
2026.06
35.1
LoopCoder-v2
R (number of refinemen...
2026.06
34.5
Qwen2.5-Coder-32B-Instruct
2026.06
32.5
DeepSeek-Coder-V2-Lite-Instruct
2026.06
26.7
Feedback
Search any
task
Search any
task