Task
Text Generation
N
nvidia/Nemotron-Labs-Diffusion-8B-Base
Text Generation·nvidia· 504.6K· 6
transformers
other
8.5B params
license:otherregion:us
Nemotron-Labs-Diffusion is a tri-mode language model that supports both AR decoding and diffusion-based parallel decoding by simply switching the attention pattern of the same model during inference. The synergy between these two modes enables a third mode, called self-speculation: the same model performs diffusion-based parallel drafting and AR verification with shared KV cache, achieving high ac
# pull & run locally
pip install mlforge-sdk && mlforge pull nvidia/Nemotron-Labs-Diffusion-8B-Base
pip install mlforge-sdk && mlforge pull nvidia/Nemotron-Labs-Diffusion-8B-Base
Model details
Provider
nvidia
Framework
transformers
Parameters
8.5B
Size
47 GB
License
other
Downloads
504.6K
Likes
6
Updated
2026-06-03
About nvidia/Nemotron-Labs-Diffusion-8B-Base
Nemotron-Labs-Diffusion is a tri-mode language model that supports both AR decoding and diffusion-based parallel decoding by simply switching the attention pattern of the same model during inference. The synergy between these two modes enables a third mode, called self-speculation: the same model performs diffusion-based parallel drafting and AR verification with shared KV cache, achieving high acceptance lengths and decoding efficiency. The seamless mode switching by simply changing attention patterns enables high efficiency at different concurrency levels in varying deployment scenarios with one single model.