Claude Code cuts first-token latency up to 65%, then a data scientist makes Claude redo the hype chart
josh_wills · x · 2026-10-18
Anthropic says Claude Code now gzips the conversation sent each turn and starts cloud sessions as you type, cutting p90 wait-before-first-word from 17.0s to 6.0s for Code cloud (-65%), 4.1s to 1.3s local (-67%), 3.4s to 2.2s for Chat & Cowork (-35%), and 1.5s to 1.0s for API (-34%, median including model time).
Data scientist vboykis reposted the announcement calling the chart "clear but containing zero information," then had Claude rework it in about 15 minutes, flagging LLM-style tells: hype-first headline, uniform template, symbol-heavy metadata, unexplained p90 jargon, and the API non-comparability caveat buried in the faintest text.
Related event: Anthropic Speeds Up Claude Code and API Responses(4 posts)→
More from coding & agent
- Grok Build v1.0.63 adds multi-agent visibility, MEMORY.md index and shell-free structured hooks — XFreeze · 2026-10-18
- OpenCode confirms breach, stolen customer API keys already incurring charges — jedisct1 · 2026-10-18
- AnyPS5 runs PS5 games on PC without an emulator, 27.8k GitHub stars — Vjeux · 2026-10-18
- JevAgentHarness: 4B Decision Model Cuts Agent Cost 14.2% With Zero Task Failures — dair_ai · 2026-10-18
- Free model Space Bunny Max proposes new best bound for integer multiplication — IgorCarron · 2026-10-18
- Ex-Citadel COO analysis: waiting for GPT-7 to agentify your company is too late — ChanceKelch · 2026-10-18