An Open Recipe for IMO Gold: Training Nemotron for Olympiad Mathematics
Open post-training pipeline turns Nemotron 3 Ultra checkpoints into an IMO 2026 gold-medal system, scoring 30/42 without formal provers or external tools.
Starting from Nemotron 3 Ultra, researchers trained two specialist checkpoints using supervised fine-tuning and reinforcement learning for natural-language olympiad proof generation. Three checkpoints power an iterative generate-verify-refine search plus a separate high-compute selection stage, operating entirely in natural language with no formal prover, external tools, or internet access. The system scored 30 of 42 points at IMO 2026, reaching the gold-medal threshold. The release includes the post-trained checkpoints, training data, training and inference code, submitted solutions, and Nemotron-IMO-Bench with 200 novel olympiad-level problems.
- Two specialist checkpoints trained via SFT and RL from Nemotron 3 Ultra
- Iterative generate-verify-refine search plus high-compute final selection stage
- 30/42 at IMO 2026 reaches the gold-medal threshold in natural language
- Releases checkpoints, data, code, solutions, and 200-problem Nemotron-IMO-Bench
- Tags
- mathematical-reasoningnemotronolympiad-mathematicsopen-weightsreinforcement-learningtest-time-compute
- Vendors
- NVIDIA
- Products
- Nemotron-IMO-Bench
- Organizations
- NVIDIA
- AI models
- Nemotron 3 Ultra
From the original article
1 min read at huggingface.co
We study how model post-training and test-time inference design affect natural-language proof generation for hard olympiad mathematics. Starting from Nemotron 3 Ultra, we train two specialist checkpoints using supervised fine-tuning and reinforcement learning, and evaluate checkpoint choice, verification, and refinement. Based on these findings, we present an open-model test-time-compute pipeline.
By Ivan Moshkov, Stephen Ge, George Armstrong · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.