nvidia
/

Nemotron-3-Nano-Omni-30B-A3B-Reasoning-NVFP4

NemotronH_Nano_Omni_Reasoning_V3

feature-extraction

8-bit precision

Model card Files Files and versions

Resources

View closed (3)

`modeling_nemotron_h.py` missing `supports_gradient_checkpointing = True` (same as BF16/FP8)

#7 opened 13 days ago by

Process to get this running on a DGX Spark with vLLM

#6 opened 14 days ago by

Disappointed: English-only support for an omni model in 2026?

#5 opened 21 days ago by