Reflection Beam Preview 2026-10-05 - tau3 banking: leaderboard
Oct5 prerelease evaluation, publisher says final red-teaming ongoing and weights forthcoming. Beam identity is date-qualified and never merged with final Oct16. Comparator exact printed model labels retained; effort/precision unreported. Public split and context-management qualifiers retained. NR cells explicitly pinned missing; a populated missing cell requires review. Metric: Score (%). Higher is better. Source: https://reflection.ai/blog/introducing-beam
Metric: Score (%). Source: reflection.ai. Saturation forecast: Around May 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.8 Max | 55.2 | #31 |
| 2 | GLM-5.2 | 37.1 | #71 |
| 3 | Kimi K3 | 37.1 | #29 |
| 4 | Inkling | 25 | #123 |
| 5 | Nemotron 3 Ultra | 22.6 | #240 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=reflection-beam-preview-2026-10-05-tau3-banking · How It Works · Data refreshed daily, snapshot 2026-10-17.