Mini toolsNemotron Labs Diffusion 14B
By NVIDIA
Model family: Nemotron
Nemotron-Labs-Diffusion-14B is a NVIDIA Nemotron language model released under the NVIDIA Nemotron Open Model License. It is a text-generation model built around tri-mode inference: standard autoregressive decoding, diffusion-based parallel decoding, and self-speculation where diffusion drafts tokens and autoregressive decoding verifies them using a shared KV cache. NVIDIA presents the Nemotron-Labs-Diffusion family as focused on decoding efficiency, with 3B, 8B, and 14B dense variants, including base, instruct, and vision-language variants.
Overview
Nemotron-Labs-Diffusion-14B is NVIDIA’s open 14B text-generation language model that supports autoregressive decoding, diffusion-style parallel decoding, and self-speculative decoding.
About NVIDIA
Industry:
Computer Hardware Manufacturing
Company Size:
42000
Location:
Santa Clara, California, US
Website:
nvidia.com
Last updated: July 21, 2026
