Hazy Research's open-source NCM generates text 31x faster via coarse-to-fine concept modeling
simran_s_arora · x · 2026-10-20
Hazy Research released Next-Concept Modeling (NCM), fully open source: a multiscale residual VQ-VAE compresses text into hierarchical discrete concepts, and a transformer predicts the hierarchy coarse-to-fine with all positions parallel — fixed 12 forward passes regardless of length. It's 10–31x faster than diffusion text models, needs 3–5x fewer training tokens, and solves hard Sudoku at 86.1% in 12 parallel steps, beating autoregressive and diffusion baselines. The reposter ran both demos on CPU in seconds.
More from Models
- Private evals: pplx-decider shows 40% recall gain over Jev decision models — marktenenholtz · 2026-10-20
- DeepSeek V4.1 refuses jailbreak persona, spotting prompt injection in red-team test — teortaxesTex · 2026-10-20
- Dev claims ~16% extra speedup for Qwen models with MTP, open llama.cpp/mlx release planned — karminski3 · 2026-10-20
- Blogger claims DeepSeek's unreleased model nearly matches Opus at design, shares one-prompt demo — sven_ai · 2026-10-20
- Fingerprint test flags DeepSeek ds-4.1-flash as likely GPT-6-astra — sven_ai · 2026-10-20
- Intelligence Saturation: 10-Point Model Edge Now Buys Just 20% More Developer Traffic — khademinori · 2026-10-20