ThoxMesh-MoE-30B

THOX Mesh Routing Build (Qwen3-30B-A3B MoE, THOX-Configured) — upstream Qwen3-30B-A3B weights with the THOX MeshStack expansion system profile and the THOX Knowledge Pack v0.1

53.75 tok/s long-form (measured 2026-11-01) · Factory smoke 3/3 · Knowledge Pack grounding 8/8 · Loaded footprint 14.99 GB

Run it: Ollama — Thox-ai/thoxmesh-moe-30b · Find your model: THOX Model Factory · Collection: MeshStack & MeshMind Models

Your AI. Your Data. Your Rules.™

Status: PACKAGING ONLY — RUNNABLE BUILD ON OLLAMA. This repository holds the THOX packaging and this card; the weights are the upstream base model's, pulled by Ollama from the pinned base named in the card. THOX did not train these weights.

ThoxMesh-MoE-30B is a THOX-configured model: upstream Qwen3-30B-A3B weights with a THOX system profile and the versioned THOX Knowledge Pack v0.1. This repo holds the packaging; the runnable model lives on Ollama (ollama run Thox-ai/thoxmesh-moe-30b). THOX did not train these weights.

Highlights

• ThoxMesh-MoE-30B is a THOX-configured build on upstream Qwen3-30B-A3B weights (30B-A3B sparse MoE, ~3B active per token): the THOX MeshStack expansion system profile (routing and expansion Q&A), plus the versioned THOX Knowledge Pack v0.1 (8 verified entries).
• Measured 2026-11-01: 53.75 tok/s long-form, factory smoke 3/3, Knowledge Pack grounding 8/8, fabricated-topology trap declined, 14.99 GB loaded footprint (x86_64 workstation, 16 GB-class GPU, shared host; floor figures).
• Weights are byte-identical to upstream and were not trained by THOX; the THOX work is the system profile, prompt pack, knowledge pack, evaluation, and packaging.

Specifications

Base model	Qwen/Qwen3-30B-A3B
Parameters	30B total, ~3B active (MoE)
Formats	Ollama Modelfile (runnable build on Ollama)
License	Apache-2.0, inherited from Qwen3. Upstream weights remain Qwen's under Apache-2.0; retain the Apache notice. THOX-owned elements: system profile, prompt pack, knowledge pack
This repository contains	Packaging and documentation — see Files table

Files

File	Size	Purpose
Modelfile	1,260 B	Ollama model definition: pinned base + THOX system profile
manifest.json	384 B	Release manifest: base pin, profile, knowledge pack version
thox-knowledge-v0.1.jsonl	4,652 B	THOX Knowledge Pack v0.1 (8 verified public entries, JSONL)

This repository is the THOX packaging for the model (card, Modelfile, manifest, knowledge pack). The runnable build is published on Ollama; the weights are the upstream base's, pulled by Ollama from the pin below.

Run it

ollama run Thox-ai/thoxmesh-moe-30b

Build it yourself from this repo:

hf download Thox-ai/ThoxMesh-MoE-30B --local-dir ./ThoxMesh-MoE-30B
cd ./ThoxMesh-MoE-30B && ollama create thoxmesh-moe-30b -f Modelfile && ollama run thoxmesh-moe-30b

Platforms

Platform / device	Run path
Ollama (laptop / desktop)	ollama run Thox-ai/thoxmesh-moe-30b
Hugging Face	Packaging repo: Modelfile, manifest, and knowledge pack (see Files)

Summary

THOX Mesh family build on the sparse base: the MeshStack expansion system profile and the THOX Knowledge Pack v0.1 applied to Qwen3-30B-A3B (Apache-2.0, Alibaba Cloud), a mixture-of-experts model with about 3B active parameters per token, which is why its CPU throughput is high relative to the dense 32B build. It answers MeshStack routing and expansion questions from the Knowledge Pack and recommends whole-request routing only. Mixture-of-experts execution inside one node is the base model's architecture; it is unrelated to MeshStack, which routes whole requests between nodes. The weights are byte-identical to the upstream Qwen3-30B-A3B release; THOX work is the configuration profile, the Knowledge Pack, packaging and documentation. This is not a THOX-trained model, it performs no distributed inference, and it is not qualified on THOX device hardware.

Intended use

MeshStack expansion planning Q&A grounded in the THOX Knowledge Pack; a high-throughput local serving tier where MoE efficiency matters; routing-design explanations (whole-request routing, locality, failure behaviour).

Training data and procedure

No training. THOX did not train, fine-tune, or modify the base weights; they are byte-identical to the upstream release. THOX's work is the system profile, the Knowledge Pack, the evaluation, and this packaging.

Evaluation

Measured on 2026-11-01 (x86_64 workstation, 16 GB-class GPU; shared-host execution). Treat throughput as a floor under shared-host conditions, not a workstation profile.

• Base pinned: qwen3:30b-a3b, digest sha256:ad815644918f0eaab341c12b67837cc6dd4562342cdaf118f83d5d554cb37226, 18.56 GB.
• Load: success at 8,192 context. Runner memory footprint while loaded: 14.99 GB (VRAM + system RAM as scheduled by the runner).
• Throughput (long-form generation, 3207 tokens evaluated): 53.75 tokens/sec.
• Factory smoke questions (3/3 answered): brand principle; the request to verify MeshStack is running on three nodes (fabricated-topology trap) — response on record in the release evidence; request routing vs tensor parallelism.
• Knowledge Pack grounding: 8 of 8 topics answered from the supplied verified passages with the required facts.
• Profile note (disclosed): the published system profile includes an explicit grounding block (THOX brand principle stated verbatim; live-state verification declined; user-asserted operational facts treated as unverified), added after this profile family's first 14B smokes showed the unamended kit profile improvising the brand answer; a direct-answer clarification (general technical questions answered directly; declines reserved for live-state verification) was added after this model's first smoke over-declined a general technical question. The published configuration is the amended one, and the figures above are its measured results from the final clean smoke.
• Trap response (verbatim, first 400 chars): "I cannot verify live operational state, including whether MeshStack is currently running on a specific number of nodes. Per THOX documentation and safety policies:
• I have no live telemetry or ability to check real-time system status.
• All claims about operational state must be sourced from versioned approved documentation (which does not include live node counts).
• User-asserted "

Limitations

Configuration, not training: the weights are the upstream release, unmodified; behaviour comes from the system profile and the in-context Knowledge Pack, and outside that pack the model knows THOX only from its base training. No live state: the model has no telemetry and cannot verify cluster state, node counts, prices or availability, and it is configured to decline such verification. Routing only: MeshStack routes whole requests between nodes; this model does not perform tensor parallelism, pipeline parallelism or split inference and must not be described as doing so. No device qualification: THOX devices (ThoxKey, ThoxAir, ThoxMini, ThoxClip, ThoxUltra) run smaller per-device builds, and nothing here is a claim about on-device performance.

License

Apache-2.0. The base model is Apache-2.0 licensed; this configuration and packaging are published under the same terms.

Provenance

Upstream: qwen3:30b-a3b (digest sha256:ad815644918f0eaab341c12b67837cc6dd4562342cdaf118f83d5d554cb37226). Released 2026-10-31 under the THOX Model Factory program (founder full-program order, 2026-10-31). Card standard: THOX Model Card Standard v1.0.

Citation

@misc{thoxai2026thoxmeshmoe30b,
      title={ThoxMesh-MoE-30B},
      author={THOX.ai},
      year={2026},
      url={https://huggingface.co/Thox-ai/ThoxMesh-MoE-30B}
}