Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Code generation on LiveCodeBench (Score, Avg)
Loading...
57
Score
RL_RACES
46.496
49.223
51.95
54.677
Jun 10, 2026
Score
Average Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
Average Score
RL_RACES
Backbone=Qwen3-14B
2026.06
57
61.1
RL_individual
Backbone=Qwen3-14B
2026.06
56.3
60.1
Base
Backbone=Qwen3-14B
2026.06
55
58.8
RL_RACES
Backbone=DeepSeek-R1-D...
2026.06
48.8
51.3
Base
Backbone=DeepSeek-R1-D...
2026.06
47.2
48.2
RL_individual
Backbone=DeepSeek-R1-D...
2026.06
46.9
48.8
Feedback
Search any
task
Search any
task