README.md · Thox-ai/thox-cluster-code at main

Add MeshStack paper research citation with DOI

7833567 verified about 12 hours ago

9.8 kB

---
license: apache-2.0
pipeline_tag: text-generation
base_model:
- Qwen/Qwen2.5-Coder-32B-Instruct
language:
- en
tags:
- thox
- meshstack
- magstack
- ollama
- gguf
- distributed-inference
- multi-device
- qwen2.5-coder
- code-generation
- software-development
- local-inference
- edge-ai
- thoxmesh
- qwen2.5
- local-first
- code
---
<div align="center">

<img src="hero-family.png" alt="The THOX four-device family" width="100%">

<h1>thox-cluster-code</h1>

**THOX configuration of Qwen2.5-Coder-32B-Instruct — a THOX configuration and distribution of Qwen2.5-Coder 32B for local software-engineering work and MeshStack deployments**

**Live on Ollama**

</div>

| ![License](https://img.shields.io/badge/License-Apache_2.0-05A451) | ![Params](https://img.shields.io/badge/Params-32B-05A451) | ![Base](https://img.shields.io/badge/Base-Qwen2.5--Coder--32B--Instruct-05A451) | ![Access](https://img.shields.io/badge/Access-Public-05A451) |
|---|---|---|---|

**Run it:** [Ollama — Thox-ai/thox-cluster-code`](https://ollama.com/Thox-ai/thox-cluster-code`) · **Find your model:** [THOX Model Factory](https://www.thox.ai/models) · **Collection:** [Code Models](https://huggingface.co/collections/Thox-ai/thox-code-models-6ae7518c544bc6b0fd740cd4)

<div align="center">
<a href="https://www.kickstarter.com/projects/thox/four-devices-one-private-ai-ecosystem-thoxai?ref=4tzhm0"><img src="https://huggingface.co/spaces/Thox-ai/README/resolve/main/back-thox-on-kickstarter.jpg" alt="Back THOX on Kickstarter" width="520"></a>
</div>

> **Your AI. Your Data. Your Rules.™**

> **Status: MODEL CARD PAGE — RUNNABLE BUILD ON OLLAMA.** This Hugging Face repository is the model's public model card page and THOX packaging record; the runnable build is served from Ollama (see Run it / How to Use). THOX did not train these weights; THOX's work is the configuration, clustering layer, and documentation.

## Summary

A THOX configuration and distribution of Qwen2.5-Coder 32B for local software-engineering work and MeshStack deployments. Configuration-derived: THOX did not train these weights. Qwen2.5-Coder is Alibaba's code-specialized model family; per Alibaba's model record it covers 92 programming languages.

> **Correction notice.** This card previously described the model as "GPT-4o competitive" and quoted an Aider score and team ROI figures; those were inherited catalog claims with no THOX-run evaluation behind them and have been removed. No benchmark, throughput, latency, concurrency, uptime, cost-saving, or compliance figure is claimed for this release. Earlier versions of this card also stated the license as MIT; that was incorrect (see License Terms).

**Model ID:** Thox-ai/thox-cluster-code
**Base model:** Qwen2.5-Coder 32B by Alibaba Cloud
**Runnable build:** Ollama - `ollama run Thox-ai/thox-cluster-code`
**License:** Apache License 2.0 (see License Terms)


## Highlights

- A THOX configuration and distribution of Qwen2.5-Coder 32B for local software-engineering work and MeshStack deployments.

## Files

| File | Size | Purpose |
|---|---:|---|
| `.gitattributes` | 1,519 B | Repository file |
| `README.md` | 7,711 | This model card |


## MeshStack Origin

This model was designed for MeshStack, THOX's device-to-device clustering
technology, built for the flagship ThoxNova devices. ThoxNova hardware was
designed to magnetically connect and interact, using custom-developed
magnetic pogo pins (MagStack) to distribute models and share compute,
memory, and workload from device to device. One device is a computer; a
stack of them is a cluster.

That physical stack is the foundation of the MeshStack app and the
patent-pending over-air mesh technology THOX ships today, which performs
the same device-to-device distribution and resource sharing without a
physical connection. MeshStack is included in the THOX Kickstarter
campaign (thox.ai).

## What THOX Did and Did Not Do

These are THOX-configured model distributions, and the THOX work lives in
the MeshStack clustering layer, configuration, documentation, and the
origin program, never in the weights. The weight layers in this model's
Ollama distribution are byte-identical to the upstream provider's Ollama
distribution (SHA-256 verified in the 2026-10-31 release review), and the
weights remain governed by the upstream license terms recorded on this
card. What THOX built is the layer around them: the MagStack and
MeshStack clustering design these configurations were made for, the
runtime configuration, and this documentation.
## Specifications

| Specification | Details |
|---|---|
| Base model | Qwen2.5-Coder 32B (Alibaba Cloud) |
| Parameters | 32B |
| Size | 20 GB (Ollama listing) |
| Deployment | Local inference via Ollama; distributed orchestration via MeshStack (separate from this download) |
| Context window | 128K tokens (served num_ctx 131072 in the Ollama configuration; the Ollama latest-tag figure of 32K is a display artifact) |
| Quantization | Q4_K_M (GGUF) |
| Languages | 92 programming languages (per Alibaba's model record) |
## Intended Use
- Software teams analysing repositories and reviewing code locally
- Startups and DevOps teams prototyping and working with infrastructure-as-code
- QA and test-automation assistance on private codebases
## How to Use
```bash
ollama pull Thox-ai/thox-cluster-code
ollama run Thox-ai/thox-cluster-code
```
This Hugging Face repository is the model's public model card page. The
runnable build is served from Ollama at
https://ollama.com/Thox-ai/thox-cluster-code - the original card this page is
based on lives there in full.
## Platforms
| Platform / device | Run path |
|---|---|
| Ollama (laptop / desktop) | `ollama run Thox-ai/thox-cluster-code`` |
| Hugging Face | Download the builds listed under Files from this repo (see Run it) |
## Training Data and Procedure
This is a THOX distribution of the base model named above, packaged and
documented by THOX.ai for local and MeshStack/MagStack deployment. The
original THOX release materials for this model claim no separate THOX
fine-tuning run, training dataset, or training procedure, and none is
claimed here. Base-model training data and procedure are those of the
base model's provider; see the base model card for that record.
Datasets added by THOX for this release: none.
## Evaluation
Benchmark and performance figures quoted in this card were reported by Alibaba (Qwen) for the base model and are reproduced from the model's original Ollama model card (catalog-reported at original publication). They were not re-run for this Hugging Face publication.
Independent THOX evaluation of this model on this page: pending. This
section will be updated with method and results when a THOX-run
evaluation lands; until then, treat quoted figures as the original
catalog's claims, not as independently verified measurements.
Catalog notes for this model: No THOX-run evaluation has been published for this model. Catalog claims that appeared here previously (an Aider score and a GPT-4o comparison) were removed; they were inherited catalog claims with no THOX-run evaluation behind them.
## Limitations
- Specialized for software engineering; not a general-purpose chat model
- Local deployment only in this distribution; behavior and speed depend entirely on the operator's hardware and configuration
- Generated code requires human review before production use
- All processing is local by design; the model has no access to online
  information or external services unless the operator wires them in.
- Outputs can be wrong. Professional, medical, legal, financial, and
  security-relevant outputs require qualified human review.
## Hardware Reference (from the original card)
Per device: minimum 32 GB RAM, 30 GB storage, NVIDIA Jetson Orin AGX class GPU (32GB VRAM), 10 Gbps network for clustered use. Recommended 64 GB RAM, 50 GB NVMe SSD.
## Research
**THOX MeshStack: Capacity-Priced Pipeline Sharding, Grant-Gated KV Transfer and Lease-Fenced Device Roles for a Private Household Mesh**
THOX.ai Research technical report v1.0.1, 3 November 2026. Craig Ross and Tommy Xaypanya, THOX.ai LLC. DOI: https://doi.org/10.5281/zenodo.23919580 — Paper page: https://thox.ai/research/papers/meshstack — PDF: https://huggingface.co/datasets/Thox-ai/thox-whitepapers/resolve/main/meshstack/THOX_MeshStack_Research_Paper_v1.0.1.pdf
## License Terms
The base model weights in this distribution are governed by the Apache
License 2.0, as recorded in the Ollama manifest license layer for this
model. THOX claims no license over the weights beyond the upstream
terms. Where a THOX copyright notice appears on a page for this family,
it applies to THOX's own material (configuration and documentation)
only, not to the weights.
## Provenance
- First listed on Ollama under Thox-ai in 2025 and held private.
- Publicly released 2026-10-31 by order of Tommy Xaypanya, CTO and
  Co-Founder of THOX.ai, as part of the MeshStack model family release.
- Runnable build: Ollama, `ollama run Thox-ai/thox-cluster-code` and
  https://ollama.com/Thox-ai/thox-cluster-code
- This Hugging Face repository was created 2026-10-31 as the model's
  public model card page. It contains this card; the runnable model
  build is served from Ollama.
- Announcement: the MeshStack model family public release announcement
  (thox.ai).
## Citation
```bibtex
@software{thoxai_thox_cluster_code_2025,
  title = {Thox.ai Cluster Code - Software Engineering MagStack Distributed AI},
  author = {Thox.ai LLC},
  year = {2025},
  url = {https://huggingface.co/Thox-ai/thox-cluster-code},
  note = {Runnable build: https://ollama.com/Thox-ai/thox-cluster-code}
}
```
**Thox.ai LLC - Frontier AI. Complete Privacy. Your Infrastructure.**
**Your AI. Your Data. Your Rules.™**

	---
	license: apache-2.0
	pipeline_tag: text-generation
	base_model:
	- Qwen/Qwen2.5-Coder-32B-Instruct
	language:
	- en
	tags:
	- thox
	- meshstack
	- magstack
	- ollama
	- gguf
	- distributed-inference
	- multi-device
	- qwen2.5-coder
	- code-generation
	- software-development
	- local-inference
	- edge-ai
	- thoxmesh
	- qwen2.5
	- local-first
	- code
	---
	<div align="center">

	<img src="hero-family.png" alt="The THOX four-device family" width="100%">

	<h1>thox-cluster-code</h1>

	THOX configuration of Qwen2.5-Coder-32B-Instruct — a THOX configuration and distribution of Qwen2.5-Coder 32B for local software-engineering work and MeshStack deployments

	Live on Ollama

	</div>

	\| ![License](https://img.shields.io/badge/License-Apache_2.0-05A451) \| ![Params](https://img.shields.io/badge/Params-32B-05A451) \| ![Base](https://img.shields.io/badge/Base-Qwen2.5--Coder--32B--Instruct-05A451) \| ![Access](https://img.shields.io/badge/Access-Public-05A451) \|
	\|---\|---\|---\|---\|

	Run it: [Ollama — Thox-ai/thox-cluster-code`](https://ollama.com/Thox-ai/thox-cluster-code`) · Find your model: [THOX Model Factory](https://www.thox.ai/models) · Collection: [Code Models](https://huggingface.co/collections/Thox-ai/thox-code-models-6ae7518c544bc6b0fd740cd4)

	<div align="center">
	<a href="https://www.kickstarter.com/projects/thox/four-devices-one-private-ai-ecosystem-thoxai?ref=4tzhm0"><img src="https://huggingface.co/spaces/Thox-ai/README/resolve/main/back-thox-on-kickstarter.jpg" alt="Back THOX on Kickstarter" width="520"></a>
	</div>

	> Your AI. Your Data. Your Rules.™

	> Status: MODEL CARD PAGE — RUNNABLE BUILD ON OLLAMA. This Hugging Face repository is the model's public model card page and THOX packaging record; the runnable build is served from Ollama (see Run it / How to Use). THOX did not train these weights; THOX's work is the configuration, clustering layer, and documentation.

	## Summary

	A THOX configuration and distribution of Qwen2.5-Coder 32B for local software-engineering work and MeshStack deployments. Configuration-derived: THOX did not train these weights. Qwen2.5-Coder is Alibaba's code-specialized model family; per Alibaba's model record it covers 92 programming languages.

	> Correction notice. This card previously described the model as "GPT-4o competitive" and quoted an Aider score and team ROI figures; those were inherited catalog claims with no THOX-run evaluation behind them and have been removed. No benchmark, throughput, latency, concurrency, uptime, cost-saving, or compliance figure is claimed for this release. Earlier versions of this card also stated the license as MIT; that was incorrect (see License Terms).

	Model ID: Thox-ai/thox-cluster-code
	Base model: Qwen2.5-Coder 32B by Alibaba Cloud
	Runnable build: Ollama - `ollama run Thox-ai/thox-cluster-code`
	License: Apache License 2.0 (see License Terms)


	## Highlights

	- A THOX configuration and distribution of Qwen2.5-Coder 32B for local software-engineering work and MeshStack deployments.

	## Files

	\| File \| Size \| Purpose \|
	\|---\|---:\|---\|
	\| `.gitattributes` \| 1,519 B \| Repository file \|
	\| `README.md` \| 7,711 \| This model card \|


	## MeshStack Origin

	This model was designed for MeshStack, THOX's device-to-device clustering
	technology, built for the flagship ThoxNova devices. ThoxNova hardware was
	designed to magnetically connect and interact, using custom-developed
	magnetic pogo pins (MagStack) to distribute models and share compute,
	memory, and workload from device to device. One device is a computer; a
	stack of them is a cluster.

	That physical stack is the foundation of the MeshStack app and the
	patent-pending over-air mesh technology THOX ships today, which performs
	the same device-to-device distribution and resource sharing without a
	physical connection. MeshStack is included in the THOX Kickstarter
	campaign (thox.ai).

	## What THOX Did and Did Not Do

	These are THOX-configured model distributions, and the THOX work lives in
	the MeshStack clustering layer, configuration, documentation, and the
	origin program, never in the weights. The weight layers in this model's
	Ollama distribution are byte-identical to the upstream provider's Ollama
	distribution (SHA-256 verified in the 2026-10-31 release review), and the
	weights remain governed by the upstream license terms recorded on this
	card. What THOX built is the layer around them: the MagStack and
	MeshStack clustering design these configurations were made for, the
	runtime configuration, and this documentation.
	## Specifications

	\| Specification \| Details \|
	\|---\|---\|
	\| Base model \| Qwen2.5-Coder 32B (Alibaba Cloud) \|
	\| Parameters \| 32B \|
	\| Size \| 20 GB (Ollama listing) \|
	\| Deployment \| Local inference via Ollama; distributed orchestration via MeshStack (separate from this download) \|
	\| Context window \| 128K tokens (served num_ctx 131072 in the Ollama configuration; the Ollama latest-tag figure of 32K is a display artifact) \|
	\| Quantization \| Q4_K_M (GGUF) \|
	\| Languages \| 92 programming languages (per Alibaba's model record) \|

	## Intended Use

	- Software teams analysing repositories and reviewing code locally
	- Startups and DevOps teams prototyping and working with infrastructure-as-code
	- QA and test-automation assistance on private codebases

	## How to Use

	```bash
	ollama pull Thox-ai/thox-cluster-code
	ollama run Thox-ai/thox-cluster-code
	```

	This Hugging Face repository is the model's public model card page. The
	runnable build is served from Ollama at
	https://ollama.com/Thox-ai/thox-cluster-code - the original card this page is
	based on lives there in full.


	## Platforms

	\| Platform / device \| Run path \|
	\|---\|---\|
	\| Ollama (laptop / desktop) \| `ollama run Thox-ai/thox-cluster-code`` \|
	\| Hugging Face \| Download the builds listed under Files from this repo (see Run it) \|
	## Training Data and Procedure

	This is a THOX distribution of the base model named above, packaged and
	documented by THOX.ai for local and MeshStack/MagStack deployment. The
	original THOX release materials for this model claim no separate THOX
	fine-tuning run, training dataset, or training procedure, and none is
	claimed here. Base-model training data and procedure are those of the
	base model's provider; see the base model card for that record.

	Datasets added by THOX for this release: none.

	## Evaluation

	Benchmark and performance figures quoted in this card were reported by Alibaba (Qwen) for the base model and are reproduced from the model's original Ollama model card (catalog-reported at original publication). They were not re-run for this Hugging Face publication.

	Independent THOX evaluation of this model on this page: pending. This
	section will be updated with method and results when a THOX-run
	evaluation lands; until then, treat quoted figures as the original
	catalog's claims, not as independently verified measurements.

	Catalog notes for this model: No THOX-run evaluation has been published for this model. Catalog claims that appeared here previously (an Aider score and a GPT-4o comparison) were removed; they were inherited catalog claims with no THOX-run evaluation behind them.

	## Limitations

	- Specialized for software engineering; not a general-purpose chat model
	- Local deployment only in this distribution; behavior and speed depend entirely on the operator's hardware and configuration
	- Generated code requires human review before production use

	- All processing is local by design; the model has no access to online
	information or external services unless the operator wires them in.
	- Outputs can be wrong. Professional, medical, legal, financial, and
	security-relevant outputs require qualified human review.

	## Hardware Reference (from the original card)

	Per device: minimum 32 GB RAM, 30 GB storage, NVIDIA Jetson Orin AGX class GPU (32GB VRAM), 10 Gbps network for clustered use. Recommended 64 GB RAM, 50 GB NVMe SSD.

	## Research

	THOX MeshStack: Capacity-Priced Pipeline Sharding, Grant-Gated KV Transfer and Lease-Fenced Device Roles for a Private Household Mesh

	THOX.ai Research technical report v1.0.1, 3 November 2026. Craig Ross and Tommy Xaypanya, THOX.ai LLC. DOI: https://doi.org/10.5281/zenodo.23919580 — Paper page: https://thox.ai/research/papers/meshstack — PDF: https://huggingface.co/datasets/Thox-ai/thox-whitepapers/resolve/main/meshstack/THOX_MeshStack_Research_Paper_v1.0.1.pdf

	## License Terms

	The base model weights in this distribution are governed by the Apache
	License 2.0, as recorded in the Ollama manifest license layer for this
	model. THOX claims no license over the weights beyond the upstream
	terms. Where a THOX copyright notice appears on a page for this family,
	it applies to THOX's own material (configuration and documentation)
	only, not to the weights.
	## Provenance

	- First listed on Ollama under Thox-ai in 2025 and held private.
	- Publicly released 2026-10-31 by order of Tommy Xaypanya, CTO and
	Co-Founder of THOX.ai, as part of the MeshStack model family release.
	- Runnable build: Ollama, `ollama run Thox-ai/thox-cluster-code` and
	https://ollama.com/Thox-ai/thox-cluster-code
	- This Hugging Face repository was created 2026-10-31 as the model's
	public model card page. It contains this card; the runnable model
	build is served from Ollama.
	- Announcement: the MeshStack model family public release announcement
	(thox.ai).

	## Citation

	```bibtex
	@software{thoxai_thox_cluster_code_2025,
	title = {Thox.ai Cluster Code - Software Engineering MagStack Distributed AI},
	author = {Thox.ai LLC},
	year = {2025},
	url = {https://huggingface.co/Thox-ai/thox-cluster-code},
	note = {Runnable build: https://ollama.com/Thox-ai/thox-cluster-code}
	}
	```

	Thox.ai LLC - Frontier AI. Complete Privacy. Your Infrastructure.
	Your AI. Your Data. Your Rules.™