[AINews] Opus 5.5 is good at explainer videos

Key points

• On vision evals, @skalskip92 ranks it Anthropic’s best vision model to date: better than Fable 5 and GPT-6 Sol, worse than GPT-6 Astra, at about 60% lower cost than Fable 5.1.1
• Terminal-Bench-Science leaders: GPT-6 Astra and Opus 5.5 lead Fable 5.1 by about 20 points.1
• DOOM agent matches show Astra at 82.5% win rate, Sol fastest, Luna best wins/$.1
• Xiaomi MiMo-V2.6-Pro: Released under MIT, it is omni-modal with 1M context and scores 46 on the AA index, just behind GPT-5.6 Sol at 47.1

Sources (1)

• [1][AINews] Opus 5.5 is good at explainer videos
  Latent Space · Sep 29, 02:44 AM

  Opus 5.5 shipped this week but the vibes are overwhelmingly positive:

  On vision evals, @skalskip92 ranks it Anthropic’s best vision model to date: better than Fable 5 and GPT-6 Sol, worse than GPT-6 Astra, at about 60% lower cost than Fable 5.1.

Extractive summary: sentences quoted from the sources.