ThoxClip-9M-role

ThoxClip Device-Role Model — the device-role model for the ThoxClip persona, fine-tuned from a THOX model trained from scratch (no external base) on ThoxClip's own role corpus

GGUF: TQ1_0

Your AI. Your Data. Your Rules.™

Status: PUBLISHED — RESEARCH MODEL. The device-role model for the ThoxClip persona, fine-tuned from a THOX model trained from scratch (no external base) on ThoxClip's own role corpus. See Files for exactly what this repo contains.

Summary

ThoxClip-9M-role is the device-role model for the ThoxClip persona, fine-tuned from a THOX model trained from scratch (no external base) on ThoxClip's own role corpus. It is a BitNet b1.58 ternary model with 8,917,248 parameters (76.4% of weights in {-1, 0, +1}). Under the current device plan it runs on the paired phone or host, not on the clip. It handles local commands, names itself when asked, and defers everything else rather than inventing it.

Highlights

• ThoxClip-9M-role is the device-role model for the ThoxClip persona, fine-tuned from a THOX model trained from scratch (no external base) on ThoxClip's own role corpus.

Specifications

Files

Run it

hf download Thox-ai/ThoxClip-9M-role thoxclip-9m-role-TQ1_0.gguf --local-dir ./ThoxClip-9M-role
llama-cli -m ./ThoxClip-9M-role/thoxclip-9m-role-TQ1_0.gguf -p "Hello" -n 128

Ollama and LM Studio cannot load ternary GGUF; use llama.cpp.

Platforms

Intended use

• Device-persona conversation for ThoxClip on a paired host: local commands, identity answers, and deferral of anything beyond the device role.
• 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/ThoxMicro-1bit-9M. 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 very small model for a narrow role; it defers rather than answers anything outside local commands.
• On-device performance is not yet measured on target hardware; no throughput figure is claimed.
• Behaviour and bias reflect the training data, which is mostly English; quality in other languages is lower.

License

Apache License 2.0. This model is fine-tuned from Thox-ai/ThoxMicro-1bit-9M, a THOX model trained from scratch and also released under Apache-2.0. Training data note: training included TinyStories, distributed under CDLA-Sharing-1.0; that license governs the dataset itself.

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

Provenance

Published by THOX.ai as Thox-ai/ThoxClip-9M-role. Upstream base: Thox-ai/ThoxMicro-1bit-9M. 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{thoxai2026thoxclip9mrole,
      title={ThoxClip-9M-role},
      author={THOX.ai},
      year={2026},
      url={https://huggingface.co/Thox-ai/ThoxClip-9M-role}
}