Anthropic launches Claude Sonnet 5.5: 30% faster, up to 30% cheaper, 70.6% on Terminal-Bench
addyosmani · x · 2026-09-29
Anthropic has launched Claude Sonnet 5.5, per employee Addy Osmani: 30% faster and up to 30% cheaper than Sonnet 5 for most workloads, targeting well-scoped everyday tasks like bug fixes, docs, and slides.
Benchmarks show clear gains in agentic coding and computer use:
- 70.6% on Terminal-Bench 4.0
- 80.1% on OSWorld 2.1
The team also showcased the generational jump by having models paint the same photo in code.
Related event: Anthropic Launches Claude Sonnet 5.5: 30% Faster, Up to 30% Cheaper(38 posts)→
More from coding & agent
- Running Codex for 5 days to enumerate every published AI safety idea — AaronBergman18 · 2026-09-29
- Sonnet 5.5 Lands in Conductor, and the Benchmark Chart Is Surprising — charlieholtz · 2026-09-29
- Agents With a Tailscale-Like VPN? The Coming Cat-and-Mouse Over Agent Browsing — alexisgallagher · 2026-09-29
- Dev benchmarks agent harnesses, says opencode beats PI, OpenClaw and Hermes by a huge margin — dh7net · 2026-09-29
- Add a support agent to your app in 5 minutes by reusing existing APIs as tools — CShorten30 · 2026-09-29
- Feed a list of easily misheard project jargon to your voice transcription model — lucasmeijer · 2026-09-29