"The more I use Opus 5.5, the more I realize how far behind OpenAI models are"
marclou · x · 2026-10-08
Developer Marc Lou says the more he uses Claude Opus 5.5, the more he feels OpenAI models have fallen behind — "not just coding. Everything."
His example: feeding GPT-6 Astra his VO2 max lab results yielded 200 sentences of "your report shows this, I think that, therefore it's inconclusive," while Opus 5.5 gave the gist and actionable steps. He likens OpenAI models to "that overthinking guy who can't make a decision" and wonders how the gap widened so much in recent months.
More from Models
- Dev ditches OpenClaw for Grok bot, citing speed and free live X access — haltakov · 2026-10-08
- Chart of AI math progress is outdated a day after OpenAI's new proofs — willknight · 2026-10-08
- Developers fume as Google sits on Gemini 4 Argon, early access hinted — OfficialLoganK · 2026-10-08
- Meta's open-weight Glimmer runs local agents at 50 tok/s on a MacBook — TimDarcet · 2026-10-08
- Glimmer model details: full training pipeline and 50 tok/s on a MacBook via 4-bit quant + DFlash — TimDarcet · 2026-10-08
- Unreleased OpenAI model produced nearly all research results from a single agent prompt — MikePFrank · 2026-10-08