Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Coding on Terminal-Bench 2.0 (Success Rate and Token Usage)
Loading...
65.2
Success Rate (SR)
HarnessBridge
16.112
28.856
41.6
54.344
Jun 11, 2026
Success Rate (SR)
Avg Input Tokens
Updated 1mo ago
Evaluation Results
Method
Method
Links
Success Rate (SR)
Avg Input Tokens
HarnessBridge
Generator Model=Claude...
2026.06
65.2
0.19
Terminus 2
Generator Model=Claude...
2026.06
64
0.26
Terminus 2
Generator Model=DeepSe...
2026.06
57.3
1.02
HarnessBridge
Generator Model=DeepSe...
2026.06
57.3
0.94
Terminus 2
Generator Model=GPT-5.4
2026.06
53.9
9.41
HarnessBridge
Generator Model=GPT-5.4
2026.06
53.9
0.99
HarnessBridge
Generator Model=DeepSe...
2026.06
53.9
1.22
Terminus 2
Generator Model=DeepSe...
2026.06
49.4
2.18
HarnessBridge
Generator Model=GPT-5....
2026.06
22.5
0.91
Terminus 2
Generator Model=GPT-5....
2026.06
18
9.8
Feedback
Search any
task
Search any
task