Infrathon Mean Rank Score - Training Module: leaderboard
Infrathon Mean Rank Score - Training Module: Mean per-task rank score (0-100). Infrathon arXiv 2611.11995v1 Figure 5: source-printed mean rank scores (0-100, higher better), not task raw score. Six panels: training operator 3, module 2, system 5; inference operator 8, module 3, system 4 tasks. Best valid private-test score within budget is ranked per task; operator 12h, module/system 24h. Claude Code for Opus 5.5, Codex CLI for the five others; source says max reasoning for all. Exact primary SVG bytes and companion paper bytes pinned. Dataset card HTTP 401 does not erase the public paper results. Native Flash identity preserved; no dated backend inferred.
Metric: Mean per-task rank score (0-100). Source: arxiv.org. Saturation forecast: Around February 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra (Max) | 90 | #6 (GPT-6 Astra) |
| 2 | Claude Opus 5.5 (Max) | 90 | #5 (Claude Opus 5.5) |
| 3 | GPT-6.1 Sol (Max) | 60 | #10 (GPT-6.1 Sol) |
| 4 | Qwen 3.8 Flash (Max) | 40 | #109 (Qwen 3.8 Flash) |
| 5 | DeepSeek V4.1 Flash (Max) | 10 | #58 (DeepSeek V4.1 Flash) |
| 6 | GLM-5.3 Flash (Max) | 10 | #119 (GLM-5.3 Flash) |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=infrathon-mean-rank-score-training-module · How It Works · Data refreshed daily, snapshot 2026-11-12.