Terminal-Bench-Science 0.1

Latest developments

Claude Fable 5.1 made me a really nice animated pelican Simon Willison’s Weblog · 2 days ago · 5 Anthropic Releases Claude Fable 5.1 and Claude Mythos 5.1: 52.6% on Terminal-Bench-Science and 75% Cheaper Cache Reads MarkTechPost · 2 days ago · 6

2026

◆ Anthropic launched Claude Fable 5.1, a developer-tools offering built around its Claude foundation models for coding and knowledge work Product launch

◆ Anthropic released Claude Fable 5.1 and Claude Mythos 5.1, including changes to agent API behavior and watermark-based text detection. Model release

• 1 Sep Claude Fable 5.1 made me a really nice animated pelican
• 1 Sep Anthropic Releases Claude Fable 5.1 and Claude Mythos 5.1: 52.6% on Terminal-Bench-Science and 75% Cheaper Cache Reads

Relationships

Products & technology

• Claude Fable 5.1 deploys this benchmark · 1 source
• Opus 5 deploys this benchmark · 1 source
• GPT-5.6 Sol deploys this benchmark · 1 source
• Claude Fable 5.1 integrated with this benchmark · 1 source