GGUF layer package for running Qwen3.5-35B-A3B-UD-Q4_K_XL across a local Mesh LLM cluster.
This package is derived from unsloth/Qwen3.5-35B-A3B-MTP-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/Qwen3.5-35B-A3B-MTP-GGUF |
| Model id | unsloth/Qwen3.5-35B-A3B-MTP-GGUF:UD-Q4_K_XL |
| Family | Qwen3 |
| Parameter scale | 35B-A3B |
| Quantization | UD-Q4_K_XL |
| Layer count | 41 |
| Activation width | 2048 |
| Package size | 21.7 GB |
| Source file | Qwen3.5-35B-A3B-UD-Q4_K_XL.gguf |
| Package repo | meshllm/Qwen3.5-35B-A3B-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/Qwen3.5-35B-A3B-MTP-GGUF.
Quickstart
# Run this on each machine that should contribute memory/compute.
mesh-llm serve --model "meshllm/Qwen3.5-35B-A3B-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/Qwen3.5-35B-A3B-MTP-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/Qwen3.5-35B-A3B-MTP-GGUF@main/Qwen3.5-35B-A3B-UD-Q4_K_XL.gguf |
| Source revision | main |
| Source SHA-256 | 26445264c88d8bd4bf465763962b222cc8f816f9abe3717421f916ec4661fa9e |
| Skippy ABI | 0.1.25 |
| Package manifest SHA-256 | bac69a30665539535a4018006397ed257d383f96909d991d7a26c6eb3c54d54c |
What Is Included
| Artifact | Path | Contents | SHA-256 |
|---|---|---|---|
| Manifest | model-package.json | Package schema, source identity, checksums | bac69a30665539535a4018006397ed257d383f96909d991d7a26c6eb3c54d54c |
| Metadata | shared/metadata.gguf | 0 tensors, 10.4 MB | 4f273c88bb5a99406fff9ad1cceba67129543a74c105cddc35e06bf17625bf6d |
| Embeddings | shared/embeddings.gguf | 1 tensors, 525.7 MB | 9c70bab4392233778a4cb40a7c8a2bc243a8561169f7f2d36afc02ca63ce76fb |
| Output head | shared/output.gguf | 2 tensors, 525.8 MB | cecb0897c2a13b151c3a20ade3a1fc7aba0a11a6711d2403d8328ac903fe2ee9 |
| Transformer layers | layers/layer-*.gguf | 41 layer artifacts, 750 tensors, 20.7 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/Qwen3.5-35B-A3B-UD-Q4_K_XL.gguf" --out-dir "/tmp/meshllm-layer-job-meshllm_Qwen3.5-35B-A3B-UD-Q4_K_XL-layers-193/package"
Links
- Source model: unsloth/Qwen3.5-35B-A3B-MTP-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