GGUF layer package for running granite-4.1-30b-UD-Q4_K_XL across a local Mesh LLM cluster.
This package is derived from unsloth/granite-4.1-30b-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/granite-4.1-30b-GGUF |
| Model id | unsloth/granite-4.1-30b-GGUF:UD-Q4_K_XL |
| Family | granite |
| Parameter scale | 30b |
| Quantization | UD-Q4_K_XL |
| Layer count | 64 |
| Activation width | 4096 |
| Package size | 16.7 GB |
| Source file | granite-4.1-30b-UD-Q4_K_XL.gguf |
| Package repo | meshllm/granite-4.1-30b-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/granite-4.1-30b-GGUF.
Quickstart
# Run this on each machine that should contribute memory/compute.
mesh-llm serve --model "meshllm/granite-4.1-30b-UD-Q4_K_XL-layers" --split
# Check the mesh and discover the OpenAI-compatible model name.
curl -s http://localhost:3131/api/status
curl -s http://localhost:3131/v1/models
# Send an OpenAI-compatible chat request.
curl -s http://localhost:3131/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{
"model": "unsloth/granite-4.1-30b-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/granite-4.1-30b-GGUF@main/granite-4.1-30b-UD-Q4_K_XL.gguf |
| Source revision | main |
| Source SHA-256 | 73bd03203cf2cf657826543d35aae5574e9674d84851345c88339a32c4ac7595 |
| Skippy ABI | 0.1.24 |
| Package manifest SHA-256 | f0c0b2f4690970fdb06ee3ded8d019afb2ab69cf1a9234695c95a181837af941 |
What Is Included
| Artifact | Path | Contents | SHA-256 |
|---|---|---|---|
| Manifest | model-package.json | Package schema, source identity, checksums | f0c0b2f4690970fdb06ee3ded8d019afb2ab69cf1a9234695c95a181837af941 |
| Metadata | shared/metadata.gguf | 0 tensors, 3.4 MB | d44f8877cc7893531cc5ace60ff512ffc0c6f3e60fea643a79b38512af9e0243 |
| Embeddings | shared/embeddings.gguf | 1 tensors, 324.9 MB | 87c599376b9a624b3eb9d2103f837f3149cdd8067bd8998a202bc1e61a1cb0c0 |
| Output head | shared/output.gguf | 1 tensors, 3.4 MB | a035991e34322996e02ce8afe0967a893d732b324cf7f2fcd2125cf1a9027db8 |
| Transformer layers | layers/layer-*.gguf | 64 layer artifacts, 576 tensors, 16.4 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.
skippy-model-package write-package "/source/granite-4.1-30b-UD-Q4_K_XL.gguf" --out-dir "/tmp/meshllm-layer-job-meshllm_granite-4.1-30b-UD-Q4_K_XL-layers-192/package"
Links
- Source model: unsloth/granite-4.1-30b-GGUF
- Mesh LLM website: meshllm.cloud
- Mesh LLM: github.com/Mesh-LLM/mesh-llm
- Discord: discord.gg/rs6fmc63eN
- Package catalog: meshllm/catalog
- Package format: layer-package-repos.md