NVIDIA open-sources Nemotron-Labs-Sigma: continuous diffusion LMs with 4-step distillation
ArashVahdat · x · 2026-11-17
NVIDIA released an update to Nemotron-Labs-Sigma, a family of open-weight language models bringing continuous diffusion to language modeling: 14 checkpoints at 3B/8B, 16-D smooth ODE trajectories enabling distillation and steering (8 TPFs for math, 4 TPFs for code), strong distillation with just 4 sampling steps, and 3.5× faster speculative decoding. All checkpoints and code are released.
More from Models
- NVIDIA publishes full Cosmos3 collection on Hugging Face — NVIDIAAI · 2026-11-17
- NVIDIA Cosmos world models hit 20 million downloads on Hugging Face — NVIDIAAI · 2026-11-17
- Orion foundation model predicts gaze: finds 2x UX bugs, cuts computer-use latency 31% — Scobleizer · 2026-11-17
- Alibaba launches Qwen4-Max: 1M context MoE model tops Chinese models on RSI-Exam — HuaxiuYaoML · 2026-11-17
- Cua and Scale AI benchmark computer-use agents: 97% non-intrusive on Omarchy vs 40% on Ubuntu — TianbaoX · 2026-11-17
- 50 sequential edits stress test: Seedream drifts least at 36.6, Nano Banana at 74.7 — aziz4ai · 2026-11-17