Downloads · 30 days
142
49% of all-time downloads
Thox-ai/ThoxAir-16M-role
ThoxAir-16M-role is a text generation model from Thox-ai. Use it when you need the model to write or continue text. It is set up for gguf. The card lists the license as apache-2.0.
Downloads · 30 days
142
49% of all-time downloads
All-time downloads
287
Public
Repo size
77.2 MB
Likes
0
Public
Click a slice to open those files.
.safetensors31.5 MB · 49%
From the Hugging Face model README
ThoxAir Device-Role Model — the device-role conversational model for the ThoxAir persona, fine-tuned from a THOX model trained from scratch (no external base)
Safetensors weights · GGUF: Q8_0, TQ1_0
</div>Your AI. Your Data. Your Rules.™
Status: PUBLISHED — RESEARCH MODEL. The device-role conversational model for the ThoxAir persona, fine-tuned from a THOX model trained from scratch (no external base). See Files for exactly what this repo contains.
ThoxAir-16M-role is the device-role conversational model for the ThoxAir persona, fine-tuned from a THOX model trained from scratch (no external base). It is a BitNet b1.58 ternary model with 15,737,088 parameters (86.6% ternary). Under the current device plan it runs on a paired host, not on ThoxAir itself. On the device it pairs with a THOX companion triage model, ThoxMesh-Head-C6, which decides which events are worth waking this model for. Footprint: 5.51 MB of weights, about 13.9 MB resident.
| Base model | Thox-ai/ThoxMicro-1bit-16M |
| Parameters | 16M |
| Formats | Safetensors weights · GGUF: Q8_0, TQ1_0 |
| License | Apache-2.0 |
| This repository contains | Full weights and tokenizer in this repo |
| File | Size | SHA-256 |
|---|---|---|
hf/model.safetensors | 31,490,008 B | 1d97408c0f800fb15600cd9df81d816aebdea6be6a3df5eb3cd924ac71766e87 |
thoxair-16m-role-Q8_0.gguf | 17,033,504 B | 18031112cc99d76d45bcc2503974ce2dd663a0d2208a8714c91bcfbe6f948488 |
thoxair-16m-role-TQ1_0.gguf | 7,391,520 B | c94e1fae97ea170c8ee29706d09531f6c39b0c219338aa8b57ffac8b457df407 |
thoxair-16m-role.tern1 | 6,147,260 B | ed1c2e1927995758846bb6c0069e7e3c799adbf491f8eef7d3f1981f996a9259 |
hf/tokenizer.json / manifest.json / .gitattributes / hf/config.json / hf/tokenizer_config.json / hf | — | Supporting files (see repo) |
| File | Use |
|---|---|
thoxair-16m-role.tern1 | TERN1 device-runtime artifact |
thoxair-16m-role-TQ1_0.gguf | llama.cpp ternary |
thoxair-16m-role-Q8_0.gguf | llama.cpp / Ollama host use |
hf download Thox-ai/ThoxAir-16M-role thoxair-16m-role-Q8_0.gguf --local-dir ./ThoxAir-16M-role
llama-cli -m ./ThoxAir-16M-role/thoxair-16m-role-Q8_0.gguf -p "Hello" -n 128
Ternary files need a recent llama.cpp build; Ollama and LM Studio load only the Q8_0 file.
| Platform / device | Run path |
|---|---|
| Hugging Face | Download the builds listed under Files from this repo (see Run it) |
THOX fine-tune of Thox-ai/ThoxMicro-1bit-16M. 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.
No THOX GATE49 result is published for this model.
Apache License 2.0. This model is fine-tuned from Thox-ai/ThoxMicro-1bit-16M, 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.™
Published by THOX.ai as Thox-ai/ThoxAir-16M-role. Upstream base: Thox-ai/ThoxMicro-1bit-16M. 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.
@misc{thoxai2026thoxair16mrole,
title={ThoxAir-16M-role},
author={THOX.ai},
year={2026},
url={https://huggingface.co/Thox-ai/ThoxAir-16M-role}
}