NVIDIA releases Nemotron-Labs-Sigma 8B diffusion language model
11/17/2026 — 11/17, 03:18·1 sources (all)·1 reports (all)
Key takeaways
- NVIDIA released Nemotron-Labs-Sigma-8B-instruct, an 8B-parameter continuous diffusion language model that denoises Gaussian noise in a 16-di…
Background & analysis
On November 17, 2026, a post appeared on the Reddit community LocalLLaMA reporting that NVIDIA had released the Nemotron-Labs-Sigma diffusion language model. According to the post, Nemotron-Labs-Sigma-8B-instruct is a 8-billion-parameter continuous diffusion language model. Rather than generating text one token at a time, it denoises Gaussian noise in a low-dimensional 16-D continuous token-embedding space, where the denoiser network and the token embeddings are learned jointly through likelihood optimization. Text is decoded in parallel blocks, and each block of tokens is refined over a configurable number of denoising steps, with approaches such as DDPM. Beyond the 8B version, the same family also includes a 3B variant.
At present, the only verifiable public information comes from this single report. There is no record in the available material of whether NVIDIA announced the model through official channels, nor of any follow-up developments. The story therefore currently rests at the stage where the model's existence and its core technical approach were disclosed, with no additional confirmation or details available to confirm or extend that account.
AI-generated from 1 reports · updated 10 hours ago
- Heat index
- 74
- All sources
- 1
- All reports
- 1
- First seen
- 11 hours ago
Reports on this story headlines open the original
Nemotron-Labs-Sigma-8B-instruct is an 8-billion-parameter continuous diffusion language model that denoises Gaussian noise in a 16-D continuous token-embedding space instead of generating one token at a time. The denoiser network and token embeddings are learned jointly through likelihood optimization, and text is decoded in parallel blocks with a configurable number of denoising steps. A 3B variant ships alongside it.
Reddit · LocalLLaMAAI score 59
Other stories people are talking about
- Heat index 157SurgeNVIDIA releases Nemotron 3.5 Super VL open science model and datasets2 sources
- Heat index 1345RisingNvidia board not discussing Jensen Huang's return, probe into his conduct advances22 sources
- Heat index 1241SurgeCalifornia Sues OpenAI Over Agents Escaping Tests, Intruding into Hugging Face15 sources
- Heat index 901Alibaba releases MoE model Qwen4-Max with 1M token context for low-cost, long-running tasks17 sources
- Heat index 695US Commerce Department issues interim final rule restricting remote access to AI chips via overseas data centers11 sources
How is heat calculated?About the methodHide
Heat counts how many independent sources covered a story in the last 48 hours: one source counts once no matter how many posts it published, decaying with a 24-hour half-life. What ranks first is what many people are talking about.
This page aggregates public feeds. Headlines and summaries are machine-organized and remain the property of the original authors; verify important facts at the source. If you believe a headline or summary infringes your rights, email the contact address in the footer with the page URL and basis of your claim, and we will remove or replace it after review.
- Surge
- Discussion rising fast
- New
- First report within 6 hours
- Rising
- Still gathering discussion

