Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Next-token Reasoning on OMNI-MATH Medium (val)
Loading...
61.15
Accuracy (Next-token Reasoning)
LoopRPT
6.498
20.6865
34.875
49.0635
Mar 20, 2026
Accuracy (Next-token Reasoning)
Average Step Length
Updated 26d ago
Evaluation Results
Method
Method
Links
Accuracy (Next-token Reasoning)
Average Step Length
LoopRPT
Model=Ouro-2.6B, Reaso...
2026.03
61.15
4
LoopRPT
Model=Ouro-2.6B, Reaso...
2026.03
60.21
2.18
LoopRPT
Model=Ouro-1.4B, Reaso...
2026.03
58.29
4
LoopRPT
Model=Ouro-1.4B, Reaso...
2026.03
57.72
2.81
Adap.
Model=Ouro-2.6B
2026.03
57.35
3.35
Peak
Model=Ouro-1.4B
2026.03
57.28
4
Adap.
Model=Ouro-1.4B
2026.03
57.2
3.53
Peak
Model=Ouro-2.6B
2026.03
57.19
4
Vanilla
Model=Qwen3-1.7B
2026.03
32.18
-
+CoT
Model=Qwen3-1.7B
2026.03
8.6
-
Feedback
Search any
task
Search any
task
