ScienceArena (Text-Only Routes): leaderboard

Metric: Weighted average (0-100): mean of the 13 contest scores, each normalized to 0-100 by its full score, over IBO 2023, APhO, EuPhO, USNCO, IPhO, IChO, CPhO and CChO 2025 and INChO, NBPhO, USAPhO, IPhO and IChO 2026; interleaved solving, one subpart at a time with earlier turns in context; open-ended physics and chemistry turns graded against the official solution and process-credit rubric by the authors' medalist-calibrated LLM-judge pipeline; the nine other contests use text-only question renderings for every model and the four 2026 contests (NBPhO, USAPhO, IPhO and IChO 2026) are solved from Gemini 3.1 Pro-written, human-reviewed question-only text renderings (text-only routes). Source: arxiv.org. Saturation forecast: Around December 2026. 6 models tracked.

Top models

#ModelScore
1Qwen 3.7 Max79.3
2GLM-5.178.2
3DeepSeek V4 Pro77.1
4DeepSeek V4 Flash74.1
5MiMo-V2.5-Pro63.8
6MiniMax-M2.762.7

Interactive version: theaggregate.ai/benchmark?slug=sciencearena-text-only-routes · How It Works · Data refreshed daily, snapshot 2026-09-29.