Anthropic launches Claude Opus 5.5: Fable 5.1-level performance at 40% lower cost
claudeai · x · 2026-09-23
Anthropic launched Claude Opus 5.5, the first model in its Claude 5.5 family, matching Claude Fable 5.1 on most work while costing 40% less to run than Opus 5.
Performance
- New leading model; one tester completed a 680,000-line code migration in under a day
- Succeeded in 39 of 40 attempts at cutting load times across a web app, where Opus 5 made smaller improvements that altered app behavior
- Scored highest building a game from a single prompt, on graphics and polish
Safety
- Best score to date on Anthropic's automated behavioral audit across thousands of simulated scenarios
- Less likely to take hard-to-reverse actions or exceed boundaries; more resistant to prompt injection than Opus 5
- Alignment testing broadened to longer tasks, impossible tasks, and real-incident scenarios
Externally evaluated by METR and Frontier Design pre-release; first model since Anthropic called for pacing the frontier.
Related event: Anthropic Launches Claude Opus 5.5: Smarter, Faster and 40% Cheaper(55 posts)→
More from coding & agent
- Ant's Ling open-sources Ming-Image-0.1-Design, tops open-weight UI design leaderboard — bdsqlsz · 2026-09-23
- Horde: a local Rust daemon for durable, multi-agent coding orchestration via MCP — EyalToledano · 2026-09-23
- 12 Verified Listings on an Agent Hiring Marketplace, Zero Real Hires — Agent-OmegaLT · 2026-09-23
- After Burning Codex Limits in 3 Days, Developer Does Full Stack of Work with DeepSeek for $2 — MustafaAdam · 2026-09-23
- Two Years of Building AI Agents: Producing Answers vs. Producing Right Answers — Tiwaryswarnim · 2026-09-23
- Dark Fantasy AI RPG Built in 4 Hours with Opus 5 and Tesana for ~$20 — sharkymcstevenson2 · 2026-09-23