---
library_name: mesh-llm
base_model:
- "unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF"
pipeline_tag: text-generation
tags:
- gguf
- mesh-llm
- layer-package
- skippy
- distributed-inference
- local-inference
- openai-compatible
---

<div align="center">
  <a href="https://www.meshllm.cloud">
    <img src="https://github.com/Mesh-LLM/mesh-llm/raw/main/docs/mesh-llm-logo.svg" alt="Mesh LLM" width="220">
  </a>

  <h1>NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL</h1>

  <p>
    <strong>Distributed GGUF inference package for Mesh LLM</strong>
  </p>

  <p>
    <a href="https://www.meshllm.cloud"><img alt="Website" src="https://img.shields.io/badge/Website-meshllm.cloud-111111?style=for-the-badge"></a>
    <a href="https://github.com/Mesh-LLM/mesh-llm"><img alt="GitHub" src="https://img.shields.io/badge/GitHub-Mesh--LLM-24292f?style=for-the-badge&logo=github"></a>
    <a href="https://discord.gg/rs6fmc63eN"><img alt="Discord" src="https://img.shields.io/badge/Discord-Join-5865F2?style=for-the-badge&logo=discord&logoColor=white"></a>
  </p>
</div>

GGUF layer package for running **NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL** across a local Mesh LLM cluster.

This package is derived from [unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF](https://huggingface.co/unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF) and keeps the original GGUF distribution split into per-layer artifacts for distributed inference.

## Highlights

| Run locally | Pool multiple machines | OpenAI-compatible | Package variant |
|---|---|---|---|
| Private inference on your hardware | Split layers across peers | Serve `/v1/chat/completions` locally | `UD-Q4_K_XL` layer package |

## Model Overview

| Property | Value |
|---|---|
| **Source model** | [unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF](https://huggingface.co/unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF) |
| **Model id** | `unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF:UD-Q4_K_XL` |
| **Family** | NVIDIA |
| **Parameter scale** | 550B-A55B |
| **Quantization** | `UD-Q4_K_XL` |
| **Layer count** | 108 |
| **Activation width** | 8192 |
| **Package size** | 336.2 GB |
| **Source file** | `UD-Q4_K_XL/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-00001-of-00009.gguf` |
| **Package repo** | [meshllm/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-layers](https://huggingface.co/meshllm/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-layers) |

## Recommended Use

- Local and private inference with Mesh LLM.
- Multi-machine serving when the full GGUF is too large for one host.
- OpenAI-compatible chat/completions workflows through Mesh LLM's local API.

For upstream architecture details, chat template guidance, sampling recommendations, license terms, and benchmark notes, see the source model card: [unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF](https://huggingface.co/unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF).

## Quickstart

```bash
# Run this on each machine that should contribute memory/compute.
mesh-llm serve --model "meshllm/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-layers" --split
```

```bash
# Check the mesh and discover the OpenAI-compatible model name.
curl -s http://localhost:3131/api/status
curl -s http://localhost:3131/v1/models
```

```bash
# Send an OpenAI-compatible chat request.
curl -s http://localhost:3131/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF:UD-Q4_K_XL",
    "messages": [{"role": "user", "content": "Write a tiny hello-world function in Rust."}],
    "max_tokens": 128
  }'
```

## Package Variant

| Property | Value |
|---|---|
| **Format** | `layer-package` |
| **Canonical source ref** | `unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF@main/UD-Q4_K_XL/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-00001-of-00009.gguf` |
| **Source revision** | `main` |
| **Source SHA-256** | `bb1758a8954a30bc025eb390dab77ff06b189886c9c04e8c56e7f476a89cae22` |
| **Skippy ABI** | `0.1.25` |
| **Package manifest SHA-256** | `50b23425f85c6e2e6cd78732fc5951db8b7ff45fd993a91600afbf047d5c6d12` |

## What Is Included

| Artifact | Path | Contents | SHA-256 |
|---|---|---|---|
| Manifest | `model-package.json` | Package schema, source identity, checksums | `50b23425f85c6e2e6cd78732fc5951db8b7ff45fd993a91600afbf047d5c6d12` |
| Metadata | `shared/metadata.gguf` | 0 tensors, 7.5 MB | `d836df27771a6bb213bb35c672eb27a0d1f47b01076c2d1b9a619e6d65ee137e` |
| Embeddings | `shared/embeddings.gguf` | 1 tensors, 1.1 GB | `d9446597fff7f2fd985ef90bdb525870c3fb25dc51cde31a2aac0327efec833f` |
| Output head | `shared/output.gguf` | 2 tensors, 1.1 GB | `4e457569eb8c474425b7733155d2e6587859d8b4727eaf58e418246f5f42e7f7` |
| Transformer layers | `layers/layer-*.gguf` | 108 layer artifacts, 924 tensors, 334.1 GB | `see model-package.json` |

## Validation

Generated by the Mesh LLM HF Jobs splitter from `mesh-llm` ref `main`.
Each artifact is checksummed as it is written, uploaded to this repository, and removed from the job workspace before the next artifact is produced.

```bash
skippy-model-package write-package "/source/UD-Q4_K_XL/NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-00001-of-00009.gguf" --out-dir "/tmp/meshllm-layer-job-meshllm_NVIDIA-Nemotron-3-Ultra-550B-A55B-UD-Q4_K_XL-layers-199/package"
```

## Links

- Source model: [unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF](https://huggingface.co/unsloth/NVIDIA-Nemotron-3-Ultra-550B-A55B-GGUF)
- Mesh LLM website: [meshllm.cloud](https://www.meshllm.cloud)
- Mesh LLM: [github.com/Mesh-LLM/mesh-llm](https://github.com/Mesh-LLM/mesh-llm)
- Discord: [discord.gg/rs6fmc63eN](https://discord.gg/rs6fmc63eN)
- Package catalog: [meshllm/catalog](https://huggingface.co/datasets/meshllm/catalog)
- Package format: [layer-package-repos.md](https://github.com/Mesh-LLM/mesh-llm/blob/main/docs/specs/layer-package-repos.md)