Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Developer Knowledge Evaluation on DevEval
Loading...
61
Win Rate
gpt-oss-120b (high)
-0.36
15.57
31.5
47.43
Jun 8, 2026
Win Rate
Candidate Score
Reference Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Win Rate
Candidate Score
Reference Score
gpt-oss-120b (high)
Params (B)=117, reason...
2026.06
61
8.82
8.24
gpt-oss-20b (high)
Params (B)=20, reasoni...
2026.06
39
8.36
8.48
GLM-5.1
Params (B)=754, reason...
2026.06
32
8.17
8.67
Qwen3.6-35B-A3B
Params (B)=35, reasoni...
2026.06
22
7.59
8.75
MiniMax-M2.5
Params (B)=230, reason...
2026.06
17
6.82
8.86
gemma-4-E2b-it
Params (B)=5.1, reason...
2026.06
3
5.99
9.11
granite-4.0-h-small
Params (B)=32, reasoni...
2026.06
2
5.3
9.15
Feedback
Search any
task
Search any
task