Downloads · 30 days
25
5% of all-time downloads
tux-ai-chat
tuxqeq/tux-ai-chat
tux-ai-chat is a machine learning model from tuxqeq. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
A Qwen3-8B chatbot fine-tuned on tokenized PII records via QLoRA.
All-time downloads
474
Public
Repo size
17.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf17.4 GB · 100%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- May 19, 2026
- Updated
- May 27, 2026
- SHA
- 1919234a
Base models
Model card
From the Hugging Face model README
tux-ai-chat
A Qwen3-8B chatbot fine-tuned on tokenized PII records via QLoRA.
What it does
All personally identifiable information is replaced with [TYPE_hash] placeholders
(e.g. [PERSON_a1b2c3d4], [SSN_e5f6g7h8]). The model can:
- Generate synthetic tokenized records
- Answer questions about specific fields in a tokenized record
- Summarize tokenized records while preserving all placeholders
- Extract and reformat sections of tokenized records
- Hold multi-turn conversations about records
It never emits raw PII and never attempts to decode placeholders.
Quickstart (Ollama)
ollama create tux-ai-chat -f Modelfile
ollama run tux-ai-chat
Example prompts:
Generate a customer record for a healthcare professional.
What is the SSN token in this record?
[paste tokenized record]
Training details
| Setting | Value |
|---|---|
| Base model | Qwen/Qwen3-8B |
| Method | QLoRA (4-bit, r=16, alpha=32) |
| Training data | 1 000 synthetic tokenized records, ~6 000 chat examples |
| Epochs | 3 |
| Thinking mode | Disabled (enable_thinking=False) |
| Quantization | Q8_0 GGUF |
Full project
- License
- apache-2.0
- Languages
- en
- Created
- May 19, 2026
- Updated
- May 27, 2026