ThoxRoute-1.5B-GGUF

<div align="center"> <img src="hero-family.png" alt="The THOX four-device family" width="100%"> <h1>ThoxRoute-1.5B-GGUF</h1>

GGUF Build of the THOX Router — q4_K_M GGUF (~0.99 GB) of Thox-ai/ThoxRoute-1.5B, THOX's preference- and intent-routing model, merged and quantized for on-device routing with llama.cpp, Ollam

GGUF: Q4_K_M

</div>

Your AI. Your Data. Your Rules.™

Status: PUBLISHED — GGUF BUILD. The Q4_K_M GGUF (~0.99 GB) of Thox-ai/ThoxRoute-1.5B, merged and quantized for on-device routing with llama.cpp and Ollama. The model card of record is the main repository's. See Files for exactly what this repo contains.

Summary

Q4_K_M GGUF (~0.99 GB) of Thox-ai/ThoxRoute-1.5B, THOX's preference- and intent-routing model, merged and quantized for on-device routing with llama.cpp, Ollama, and LM Studio.

Chat template:

<|im_start|>system
{system}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Highlights

• Q4_K_M GGUF (~0.99 GB) of Thox-ai/ThoxRoute-1.5B, THOX's preference- and intent-routing model, merged and quantized for on-device routing with llama.cpp, Ollama, and LM Studio.
• bash hf download Thox-ai/ThoxRoute-1.5B-GGUF --include '*.gguf' --local-dir ./ThoxRoute-1.5B-GGUF llama-cli -m ./ThoxRoute-1.5B-GGUF/thoxroute-1.5b-Q4_K_M.gguf -p "Hello"

Specifications

Files

Run it

hf download Thox-ai/ThoxRoute-1.5B-GGUF --include '*.gguf' --local-dir ./ThoxRoute-1.5B-GGUF
llama-cli -m ./ThoxRoute-1.5B-GGUF/thoxroute-1.5b-Q4_K_M.gguf -p "Hello"

ollama run hf.co/Thox-ai/ThoxRoute-1.5B-GGUF:Q4_K_M

Also published in the Ollama library as thox-ai/thoxroute. Sized for devices with at least 4 GB RAM.

Platforms

Intended use

• On-device and edge intent/preference routing within the THOX stack.
• Not a general assistant and not safety-aligned for direct public exposure; deploy behind THOX guardrails on any public-facing path.
• Out of scope: medical, legal, financial, safety-critical, or other high-stakes decisions without human review; unattended control of physical devices without a confirmation step.

Training data and procedure

THOX fine-tune of Thox-ai/ThoxRoute-1.5B. The corpus and procedure are described in this card as published at release (see Summary and Evaluation); no training beyond what this card describes is claimed for this repository.

Evaluation

No THOX GATE49 result is published for this model.

Limitations

• A router, not an answerer: it selects routes; the quality of the final answer depends on the model routed to.
• Quantization to Q4_K_M can change routing decisions at the margins.
• Behaviour and bias reflect the training data, which is mostly English; quality in other languages is lower.

License

Apache-2.0. Quantized from the THOX router adapter; the THOX additions are covered under Apache-2.0. See the license text for terms.

THOX.ai — Your AI. Your Data. Your Rules.™

Provenance

Published by THOX.ai as Thox-ai/ThoxRoute-1.5B-GGUF. Upstream base: Thox-ai/ThoxRoute-1.5B. This card was conformed to THOX Model Card Standard v1.0 on 2026-10-31 (frontmatter, tags, header, files, and section structure); the technical content is as published at release.

Citation

@misc{thoxai2026thoxroute15bgguf,
      title={ThoxRoute-1.5B-GGUF},
      author={THOX.ai},
      year={2026},
      url={https://huggingface.co/Thox-ai/ThoxRoute-1.5B-GGUF}
}