Claude Code creator Boris Cherny: general models almost always beat fine-tuned ones
sull · x · 2026-09-29
Boris Cherny, creator of Claude Code at Anthropic: "The more general model will always outperform the more specific model. Don't try to use tiny models for stuff. Don't try to fine-tune." He concedes there are niche exceptions, but says developers with the flexibility should almost always bet on the general model.
Related event: Claude Code Creator: General Models Beat Fine-Tuned Small Models(2 posts)→
More from Models
- Leaked benchmarks claim Claude Sonnet 5.5 hits 56 on AA index, beating Opus 5.5 — airesearch12 · 2026-09-29
- Leaked benchmark shows Sonnet-5.5 above Astra, source unverified — adonis_singh · 2026-09-29
- Blogger predicts Chinese models will match Opus 5.5 in 2-3 months at 90% less cost — rchardkovacs · 2026-09-29
- Sonnet 5.5 Is Not a Very Good Model, Says Founder Bindu Reddy — bindureddy · 2026-09-29
- Users report surge in false-positive [cyber] Claude safeguards since Sonnet 5.5 — ofhgtl · 2026-09-29
- Asked about copyright, Gemini researched how to get an annoying person to stop talking — Nice_Assumption_6396 · 2026-09-29