Early user: GPT-6 Sol feels slower and dumber than 5.6 in Codex
burkov · x · 2026-10-01
ML practitioner Andriy Burkov reports that GPT-6 Sol feels slower and dumber than 5.6, saying it's the first time in months he feels he's wasting his time using Codex — a notable negative first-hand review of the new model's coding performance.
More from coding & agent
- At Jevathon, an AI band where Jev only conducts every 2.7s bar — schwentker · 2026-10-01
- OpenAI's new Decisions API looks a lot like Jev — and 7 hackathon teams built it first — schwentker · 2026-10-01
- How to ask a question people will actually answer: keep it under three lines — jackedAJ · 2026-10-01
- ML engineer: writing a bs=1 trainer myself beats explaining it to an LLM — cephaloform · 2026-10-01
- From orchestrating Codex threads to just asking: one engineer's simpler agent workflow — gabrielchua · 2026-10-01
- Private AI Proxy for Mac verifies AI services via hardware attestation before agent code leaves your machine — bgmshana · 2026-10-01