Feature Request: Nemotron-3-Nano-30B-A3B model (moe on nemotron_h)

[mattepiu]

Prerequisites

• I am running the latest code. Mention the version if possible as well.
• I carefully followed the README.md.
• I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• I reviewed the Discussions, and have a new and useful enhancement to share.

Feature Description

Support for Nvidia Nemotron-3-Nano-30B-A3B and architecture nemotron_h_moe which should be a moe extension of the existing nemotron_h.

Motivation

Benchmarks are very good for its size (i.e.: 38.8 in swe-bench) and speed is quite better than similar sized qwen3 models.

Possible Implementation

No response

[gabe-l-hart]

Looks like this is already addressed in #18058

[BVEsun]

b7415

llama-server.exe

unsloth/Nemotron-3-Nano-30B-A3B-GGUF/Nemotron-3-Nano-30B-A3B-UD-Q6_K_XL.gguf

llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'nemotron_h_moe'

Looks like this is already addressed in #18058

just waiting merge

Sorry for missing the above comment

[aussetg]

2995341

:)