SolsticeAI's picture
Add official Solstice-AI documentation
613dfeb verified
|
Raw
History Blame Contribute Delete
1.3 kB
---
language:
- en
- zh
license: apache-2.0
base_model: DavidAU/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored
tags:
- solstice-ai
- davidau
- qwen
- qwen3.8
- qwen3.8-27b
- cold-fusion
- gain
- project-heretic
- heretic
- uncensored
- mxfp4
- ocp
- microscaling
- vllm
- sglang
- compressed-tensors
- 4-bit
- vision
- multimodal
- mmproj
pipeline_tag: image-text-to-text
---
<p align="center">
<img src="https://cdn-uploads.huggingface.co/production/uploads/67c2e844e0921a5410eec10a/Y5M42dCag2f7Fc6fDtV0Z.jpeg" alt="Solstice-AI Banner" width="100%">
</p>
<h1 align="center">Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored (MXFP4)</h1>
<h3 align="center">Official Solstice-AI OCP Microscaling FP4 Suite</h3>
<p align="center">
<b>Original Model & GAIN Merge by <a href="https://huggingface.co/DavidAU">DavidAU</a> &bull; MXFP4 Quantization & Packaging by <a href="https://huggingface.co/Solstice-AI">Solstice-AI</a></b>
</p>
## Overview
OCP Microscaling FP4 (MXFP4) format via `compressed-tensors` with dynamic group scaling, optimized for high throughput inference across modern accelerators.
### Serving with vLLM
```bash
vllm serve Solstice-AI/Qwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-MXFP4 \
--tensor-parallel-size 1 \
--max-model-len 262144
```