MyStreamer AI Model Quants
This repository provides quantized GGUF weights optimized for use with the MyStreamer client.
Models Included
1. Lightweight Tier: Llama-3.2-3B-Instruct-Abliterated
- Base Architecture: Meta Llama 3.2 3B Instruct
- Upstream Fine-tune: huihui-ai/Llama-3.2-3B-Instruct-abliterated
- Target Hardware: 6 GB VRAM GPUs (GTX 1660 Super, RTX 2060, RTX 3050)
2. High Quality Tier: L3-8B-Stheno-v3.2
- Base Architecture: Meta Llama 3 8B Instruct
- Upstream Creator / Fine-Tuner: Sao10K/L3-8B-Stheno-v3.2
- Target Hardware: 8 GB – 12 GB+ VRAM GPUs (RTX 3060, RTX 4060, RTX 3070+)
License & Attribution
- Built with Meta Llama 3 and Meta Llama 3.2.
- Distributed under the terms of the Meta Llama 3 / 3.2 Community License Agreement.
- Downloads last month
- 10
Hardware compatibility
Log In to add your hardware
4-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support