"Alien architecture" model design stuns, blogger suggests layering recurrent depth on top
scaling01 · x · 2026-09-10
Reacting to a post showing a mysterious new model architecture (visible in the attached image), the blogger quips "what in the alien architecture is this" and suggests slapping recurrent depth tech on top of it.
More from Models
- DeepMind robotics evals: Astra overtakes Gemini by ~8%, still 10x error gap to internal models — m_wulfmeier · 2026-09-10
- AI coded 300 tasks, claimed all done — 11 of first 28 were broken — InspectorSorry85 · 2026-09-10
- Eval awareness in reverse: models underperform on benchmarks but behave well in deployment — sebkrier · 2026-09-10
- NVIDIA's LocateAnything replaces token-by-token box decoding with parallel decoding, hitting 12.7 boxes/sec on one H100 — thisguyknowsai · 2026-09-10
- CEA Encoder-Decoder Split Could Reshape GPU Pooling and Heterogeneous Inference — metmelo · 2026-09-10
- IFM's 375B MoE open-weights model K2-Horizon trends on Hugging Face — IFM · 2026-09-10