Opus 5 System Card Reveals Algebra Conversion Tactic for ARC-AGI Puzzles
herbiebradley · x · 2026-07-25
A researcher dug into the Claude Opus 5 system card, noting it performs noticeably better on ARC-AGI 1 and 2 compared to versions 4.7/4.8.
The analysis highlights that Opus 5's strategy for ARC-AGI-3-like tasks involves converting visual puzzles into explicit algebra—a highly trainable tactic. This approach is nearly identical to strategies previously used with scaffolding to solve ARC-AGI-3 tasks using Opus 4.6. Separately, Mechanize is now seeking a puzzle maker to design complex challenges that current LLMs cannot yet solve.
Related event: Claude Opus 5 Sets New ARC-AGI-3 Record(15 posts)→
More from Models
- GPT-6.1 Terra spotted in the OpenAI API, teased by benchmark group LuminaBench — airesearch12 · 2026-10-27
- Ortet's first health eval: 34 open-weight models, 50 tasks, generic rankings fall short — thegautamkamath · 2026-10-27
- Fern 0.1: a 0.1–0.5T coding model one researcher pretrained in 3.5 months for under $10M — teortaxesTex · 2026-10-27
- VLM Run launches OCR Playground for side-by-side comparison of GLM-OCR, DeepSeek-OCR-2 and more — spillai · 2026-10-27
- GPT-6.1-Terra Live With Big Price Cut: $5 vs $12 Output Pricing — scaling01 · 2026-10-27
- Nat Lambert: traces from mid-tier open models are almost all super janky — natolambert · 2026-10-27