Chaubard: HRM scored 70% on ARC Prize 1 without pre-training
Francois Chaubard · Recursion Is The Next Scaling Law In AI · Y Combinator · May 1, 2026 · at 10:09
Francois Chaubard discusses the benchmark performance of the 27M parameter Hierarchical Reasoning Model (HRM) on ARC Prize puzzles compared to OpenAI models.
0:00 / 0:23exact quote · 23.0s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“There is no pre-training at all. This starts from, like, literally Tagula-Rasa weights, and it can outperform at that time, if we go back, you know, we had O-three, if you remember back, way back when. And it, O-three gets zero. Literally zero, and this got, like, something like 70% on ArcPrize one at least at the time, which was just a huge breakthrough.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Francois Chaubard
Chaubard: Recursive models discover problem-solving strategies without human teacher forcing
“That's the most important part, is that if we had Sudoku, and we know how to solve Sudoku, because like we were just, you know, dumb homo sapiens that didn't know how to solve Sudoku, like it would just have solved it. And that's why it's cool, because it actu…”
Chaubard predicts running recursive reasoning models inside giant LLM latent spaces
“What you can imagine is we found mapping from token space or from vision, from pixels, Some really cool latent space where, like, things are just nicely semantically separated, and we can, you know, makes it really easy for downstream tasks to do, but now in t…”
Chaubard: AI models are recursively distilling into one another, Claude into Kimi
“What I think is kind of happening right now, but like clawed kind of mother birds into, Kimmy too, and then now Kimmy too is post-training thinking machines, and so it just keeps going and going.”
Chaubard: Hierarchical Reasoning Models Offer Little Novelty Over Standard RNNs
“The, this is directly in the lineage of RNNs. There's not that much novel from, like, the RNN standpoint at least in my opinion.”