Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Code Generation on MultiPL-E MBPP
Loading...
58.8
Score
Kimi-K2
50.272
52.486
54.7
56.914
Jan 6, 2026
Jan 12, 2026
Jan 18, 2026
Jan 24, 2026
Jan 30, 2026
Feb 5, 2026
Feb 11, 2026
Score
Pass@10
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
Pass@10
Kimi-K2
Model Variant=Base, #...
2026.02
58.8
-
Kimi-K2 Base
# Shots=0-shot, # Acti...
2026.01
58.8
-
Step 3.5 Flash
Model Variant=Base, #...
2026.02
58
-
MiMo-V2 Flash
Model Variant=Base, #...
2026.02
56.7
-
MiMo-V2-Flash Base
# Shots=0-shot, # Acti...
2026.01
56.7
-
DeepSeek V3.1
Model Variant=Base, #...
2026.02
52.5
-
DeepSeek-V3.1 Base
# Shots=0-shot, # Acti...
2026.01
52.5
-
DeepSeek V3.2
Model Variant=Exp Base...
2026.02
50.6
-
DeepSeek-V3.2 Exp Base
# Shots=0-shot, # Acti...
2026.01
50.6
-
Full
Training Strategy=Full...
2026.06
-
82.1
MSA-PT
Training Strategy=from...
2026.06
-
81.6
MSA-CPT
Training Strategy=spar...
2026.06
-
81.1
Feedback
Search any
task
Search any
task