Anthropic engineer: don't use Sonnet at max effort — just use Opus
edwinarbus · x · 2026-09-29
Edwin Arbus of Anthropic says it outright: don't run Sonnet at max effort. At that point you should be using Opus.
- Sonnet is the balance of quality/speed/cost; pushing it to max makes it think more, take longer, and cost more, losing what makes it useful
- That's why default effort levels exist — Claude Code defaults to medium
- The effort dials remain for edge cases, but the recommendation for most users is: don't touch that dial
Related event: Anthropic Staff: Don't Max Out Sonnet's Effort, Use Opus Instead(2 posts)→
More from coding & agent
- Continual-ARC: lifelong ARC-AGI stream finds agent harness persistence rarely beats simple caching — unifyai · 2026-11-16
- WebArena-Pro: a 300-task multimodal benchmark across 20 self-hosted apps for web agents — McGill-NLP · 2026-11-16
- Filmworkbench v0.1.2: one-prompt semi-automated short films with Claude and ComfyUI — Eastern-Ad9515 · 2026-11-16
- Stop prompting LLMs for JSON: PowerShell module Jev routes with decision models — dfinke · 2026-11-16
- llama.cpp offloads CI caches and runners to Hugging Face to survive 35 PRs/day — ggerganov · 2026-11-16
- I killed my 30-minute weekly sync with a decision log my agents read before every run — Humaux · 2026-11-16