NVIDIA open-sources Nemotron-Labs-Sigma: 14 continuous-diffusion LLM checkpoints at 3B/8B
ArashVahdat · x · 2026-11-18
NVIDIA released the Nemotron-Labs-Sigma open weights — 14 checkpoints at 3B and 8B (Sigma originated as Zhihan Yang's internship project). The key idea: bring continuous diffusion from vision to language. Smooth ODE trajectories (16-D) unlock distillation and steering, shipped as 8 math-focused and 4 code-focused TPF variants while preserving accuracy and diversity. Announced by Morteza Mardani and celebrated by NVIDIA researchers including Arash Vahdat.
More from Models
- Astra 6.2 plays detective: finds URL on a projector screen in hours of B-roll — gabrielchua · 2026-11-18
- Qwen 3.8 27B vs Flash-Next vs DeepSeek V4.1 Flash: one-shot landing page build-off — jesdga95 · 2026-11-18
- GPT-6.2 Astra plays RimWorld for 250 in-game days and builds a decent colony — jxnlco · 2026-11-18
- Polymarket claims GPT-6 Cyber uncovered 35 security flaws across Apple iPhones and Macs — Polymarket · 2026-11-18
- OpenAI ships GPT-6.2 Astra: 22-point ExploitBench jump, tighter cyber safeguards — charliermarsh · 2026-11-18
- Halocline inference engine hits 148-182 tok/s on Qwen-Flash-Next with KLD 0.082 — Distinct-Pie2389 · 2026-11-18