open source
An Open Model Reached the IMO Gold Line Without a Formal Prover
Nemotron specialists, iterative verification and a high-compute selector scored 30 of 42 points on the 2026 olympiad.
Summary
Nemotron specialists, iterative verification and a high-compute selector scored 30 of 42 points on the 2026 olympiad.
Researchers post-trained two Nemotron 3 Ultra specialist checkpoints, then combined them with the general model in an iterative generate, verify and refine pipeline. The system used natural-language proofs only—no formal prover, internet access or external tools—and scored 30 out of 42 at IMO 2026, the reported gold-medal threshold. The release includes both specialist checkpoints, training data, inference code, submitted solutions and a 200-problem benchmark. It is a high-compute recipe, so the score does not by itself establish low-cost or broadly reliable mathematical reasoning.
Why it matters
Nemotron specialists, iterative verification and a high-compute selector scored 30 of 42 points on the 2026 olympiad.
Limits and context
- The system used natural-language proofs only—no formal prover, internet access or external tools—and scored 30 out of 42 at IMO 2026, the reported gold-medal threshold.
- It is a high-compute recipe, so the score does not by itself establish low-cost or broadly reliable mathematical reasoning.
Key claims
Nemotron specialists, iterative verification and a high-compute selector scored 30 of 42 points on the 2026 olympiad.
Qualification: The system used natural-language proofs only—no formal prover, internet access or external tools—and scored 30 out of 42 at IMO 2026, the reported gold-medal threshold.
Evidence: source-2026-09-11-003
Sources
- arXiv preprint 2609.10712arXiv · primary research
Corrections
No corrections have been recorded for this story.