Why is Astra so good at 3D modeling? Better RL envs vs dropping English CoT
willdepue · x · 2026-09-07
- Hypothesis (akarvonen): Astra's big jump in 3D modeling comes from strong no-CoT capabilities — reasoning in English tokens has long seemed like a bottleneck for spatial reasoning.
- Counterpoint (willdepue): "it's probably just better RL envs honestly, I'm pretty sure."
- He adds that Astra's perception also improved markedly — likely mostly scale, but likely some pretraining architecture/encoder wins too: "Astra can see really well and this surely helps."
Related event: Debate Over Astra's 3D Leap: No-CoT Ability or Scale?(3 posts)→
More from Models
- OpenAI upgrades Decisions API: 400 questions per call, native voice and image support — OpenAIDevs · 2026-10-26
- Fangcun AI, China's first independent frontier AI safety lab, launches with DeepSeek eval report — JacquesThibs · 2026-10-26
- When AI hits an existential crisis: model confuses itself and everyone else — Ok-Delivery-7437 · 2026-10-26
- Kimi K3.1 Pro tops MathArena open models, closing frontier gap to about 6 months — teortaxesTex · 2026-10-26
- Ruff creator Charlie Marsh: frontier models find infinite bugs in any software — charliermarsh · 2026-10-26
- Dev testing one skill across Opus 5.5 and Sol 6.1 finds instruction following 'all over the map' — lucasmeijer · 2026-10-26