signal爱范儿2026-09-23
Claude 5.5 Released, Performance Rivals Fable, and Cuts Prices
Anthropic released Claude Opus 5.5, matching Claude Fable 5.1 performance on most tasks while cutting typical task costs by 40% and boosting output speed by over 30% versus Opus 5. API prices drop 20% to $4/$20 per million tokens, with cached reads down 60% to $0.20. It scores 66.4% on Terminal-Bench 4.0, surpassing GPT-6 Astra's 57.9%, and handles a 680k-line code migration in a day, with long-task costs 51% lower than Fable 5.1.
- for who
- Developers and organizations building long-running AI agents or knowledge workflows
- why now
- Opus 5.5发布，降价20%，性能直逼Fable
- what changes
- It makes near-frontier AI affordable enough to run continuously, turning advanced intelligence into a cost-effective commodity for personal agents.
- to do
- Try Claude Opus 5.5 for multi-hour coding or analysis tasks to leverage its lower cost and higher speed.
key points
- Opus 5.5 scores 66.4% on Terminal-Bench 4.0, beating GPT-6 Astra's 57.9%
- Pricing: $4 input, $20 output per million tokens, 60% cheaper cache reads
- Long tasks: 680k-line migration in a day; HAProxy rewrite in 9.5h vs Fable's 12h
#claude-opus-5.5#anthropic#model-release#performance-comparison#price-cut5 sources · confidence high
score
score 9 out of 10. 0-10: how dense the facts are, multiplied by how much you can do with them after reading. 8+ means the topic's evidence bar is met: benchmarks and availability for a new model, amount and investors for a funding round, revenue figures for a solo-money story. Below 5 an item does not enter the digest. A press release scores 3 or less, a reprint loses 2, anything older than 14 days loses 1, a headline that misleads loses 3.
read the source