Anthropic cuts cache-reading fees by 75% in Claude Fable 5.1
Anthropic announced on September 1, 2026, that its top-tier model, Claude Fable 5.1, will be up to 45% cheaper for certain use cases, though the base price remains unchanged. The cost reduction stems from a 75% drop in the cache-reading fee, which applies when the model re-reads previously processed input—such as rereading a PDF or intermediate steps in an agent task. Anthropic compared this to…
Key points
- Cache-reading fees dropped 75%, cutting costs for repeated input processing in Claude Fable 5.1
- General use sees ~25% savings; delegated tasks (agents) save up to 45% based on August 2026 data
- Safety filters now block biology/medical queries 85% less often than in Fable 5
The savings vary by usage: general tasks see about a 25% discount, while delegated tasks (e.g., AI agents handling long workflows) can save up to 45%, based on August 2026 data. Additionally, the model’s safety filters now block basic biology and medical questions 85% less frequently than in Fable 5, though complex or sensitive queries may still trigger Claude Opus 5 for handling. Anthropic’s CTO, Ogiso, emphasized that Claude Opus 5.5—released September 22—offers comparable capabilities at under half the price of Fable 5.1, suggesting users weigh cost against performance for their needs.
Model pages: Claude Fable 5.1 → · Claude Opus 5.5 →
The story so far
3 episodes →- Anthropic cuts cache-reading fees by 75% in Claude Fable 5.1this story
What is the cost of Claude Fable 5.1? Up to 45% cheaper with no price cuts [EeeeAI]
note.com · 5 October 2026
Loading the full article…
This text was published by note.com and written by イーモジャパン株式会社. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- Anthropic prompts users to share voice data to enhance AI models · 8 src
- SentinelOne engineer uses GPT-6 Astra to crack 217-year-old Napoleonic cipher · 2 src
- Reka AI releases research preview of Rho-1 omni-model · 2 src
- Reflection AI debuts Beam, a 501B open-weight model rivaling GLM 5.2 with 3-4x less compute · 7 src
- Microsoft launches MAI-Transcribe-2-Streaming real-time transcription AI for $0.54 per hour · 3 src
Comments
via GitHub Discussions