Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Autonomous Software Engineering on SWE-Bench Pro (test)
Loading...
32.6
Resolution Rate
Sonnet
29.896
30.598
31.3
32.002
May 4, 2026
Resolution Rate
Main Agent Token Usage (k)
Subagent Token Usage (k)
Frontier LLM Token Usage (k)
Main Agent Terminal Count
Subagent to Terminal Ratio
Final Answer Accuracy
Updated 2mo ago
Evaluation Results
Method
Method
Links
Resolution Rate
Main Agent Token Usage (k)
Subagent Token Usage (k)
Frontier LLM Token Usage (k)
Main Agent Terminal Count
Subagent to Terminal Ratio
Final Answer Accuracy
Sonnet
Subagent Config.=Sonne...
2026.05
32.6
756
19
774
0.6
6
100
SFT-4B
Subagent Config.=SFT-4...
2026.05
32.4
727
17
727
1.1
17
98.5
Opus
Subagent Config.=Opus,...
2026.05
32
729
19
748
0.5
4
100
Terminus-4B
Subagent Config.=Termi...
2026.05
31.5
730
18
730
1
14
97.9
Vanilla-4B
Subagent Config.=Vanil...
2026.05
30.4
840
6
840
1.9
27
99
No Subagent
Subagent Config.=No Su...
2026.05
30
836
-
836
3.8
-
-
Feedback
Search any
task
Search any
task