Claude Code and Codex Beat Human Speedrun Records, But No Model Invented a New Optimizer
AI Engineer · youtube · 2026-09-27
Prime Intellect research engineer Elie Bakouch (creator of SmolLM) set Claude Code and Codex loose on the community's Optimizer Speedrun — training a GPT-2-level model in the fewest steps. Both agents beat the human record.
Behavior differed sharply: Claude Code stopped every 9-10 hours declaring the record unbeatable and sat idle 1/3 of the time; Codex never quit, wrote far more notes, spawned more sub-agents and burned more tokens. In a six-day run, Kimi was the most token-efficient, and a paper only Claude surfaced led to the best record.
His sobering takeaway: none of the models invented a new optimizer — they combined existing ideas for small gains. He closes with an AlphaEvolve-style discovery loop Prime Intellect is building, arguing this research should happen in the open.
More from coding & agent
- Supermemory open-sources its Slack-based company brain AI teammate with one-click self-host — _AustinCalvert_ · 2026-09-27
- Dev hands an Opus 5.5 his released web game to improve and cut a trailer — hullabaloo22 · 2026-09-27
- Grok Build v1.0.41 ships subagent config inheritance and long-reasoning fixes — mark_k · 2026-09-27
- Harrison Chase on building frontier harnesses: context layers, evals, online learning — VeryWellVersed · 2026-09-27
- Dev shares public Grok Bot 'chief of staff' template running a background specialist agent swarm — RachelVT42 · 2026-09-27
- Meta pitches hand-first dev workflow: one Unity codebase across Quest and VR glasses — Scobleizer · 2026-09-27