Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Agentic on SWE-bench Verified
Loading...
73.8
Score
GLM-5
21.384
34.992
48.6
62.208
Jun 13, 2026
Jun 16, 2026
Jun 20, 2026
Jun 24, 2026
Jun 28, 2026
Jul 2, 2026
Jul 6, 2026
Score
Updated 18d ago
Evaluation Results
Method
Method
Links
Score
GLM-5
Thinking Mode=non-thin...
2026.06
73.8
Ling-2.6-1T
2026.06
72.2
GPT-5.4
Reasoning Mode=non-rea...
2026.06
69.2
Kimi-K2.5
Mode=Instant
2026.06
66.8
DeepSeek-V3.2
Thinking Mode=nothink
2026.06
66.4
Ling-2.6-flash
2026.06
61.2
Nemotron-3-Super 120B-A12B
Evaluation Setting=non...
2026.06
61
Audex 30B-A3B
Context Length=1M
2026.07
48.2
GPT-5.4-mini
Evaluation Setting=non...
2026.06
43.8
Qwen3-Omni 30B-A3B Thinking
Context Length=64K
2026.07
23.4
Feedback
Search any
task
Search any
task