competition
Improve the math reasoning capabilities of a 4B model using post-training and inference-time techniques — no external APIs, tools, or second model at inference time.
943
Maths questions
Qwen3–4B
74.5%
Pushing the boundaries of a small language model.
2nd Solo8th Team
Improve the math reasoning capabilities of a 4B model using post-training and inference-time techniques — no external APIs, tools, or second model at inference time.
COMPUTE / RECORD
reasoning tokens generated
THE DATASET
sources
SuperGPQA · UGMathBench
licence
CC BY-NC-SA 4.0
<think>
</think>