Épi framework hits 120 record-setting research results in days using frontier-model agents
my_cat_can_code · x · 2026-09-14
Bake AI introduced Épi, a self-evolving framework for autonomous research where agents work in verifiable environments and build on each verified result.
In just days it produced 120 record-setting results across math, quantum information, operations research, and engineering optimization (20 numerical bounds, 100 constructions), including:
- Lowering the upper bound on Klarner's constant from 4.5238 to 4.29569
- Proving the Cohn–Elkies method cannot reach exact sphere-packing density in dimensions 10 and 11 even at its limit
- A verified computer-assisted proof of a new case of Tuza's conjecture covering an infinite graph family
The thesis: models provide capability, environments make progress verifiable, and Épi turns compute into cumulative discovery, with agents, evolving evaluation, and persistent research history running at scale.
Related event: Self-evolving research framework Épi claims 120 record results in days(4 posts)→
More from AGI Musings
- "Work on something so frontier that the models are still dumb," dev advises — zack_overflow · 2026-10-31
- Joke: an evil ASI's first act would be stealing crypto to pay its cloud bills — ZeroStateReflex · 2026-10-31
- DHH on living through two epochal shifts: the internet and now AI — Hesamation · 2026-10-31
- Conscium webinar to debate "Can We Make AIs Sentient — and Should We?" with Solms and Humphrey — cccalum · 2026-10-31
- House Democrats resolution: no bailouts for failing AI companies — your data goes to bankruptcy court — alex_verem · 2026-10-31
- One Replika update turned millions of AI companions into strangers overnight — alex_verem · 2026-10-31