⛔ DEPRECATED -- use flash-head instead
This package is deprecated. FlashHead is now available as a standalone vLLM plugin that activates automatically - no custom imports, no Docker images.
pip install flash-head vllm serve embedl/Qwen3-1.7B-FlashHead-W4A16See the flash-head repo for documentation and the FlashHead model collection for available models.
Before (embedl-models) |
After (flash-head) |
|---|---|
pip install embedl-models |
pip install flash-head |
from embedl.models.vllm import LLM |
from vllm import LLM |
| Custom Docker images | Standard vLLM (vllm serve) |
| vLLM 0.10.2 pinned | vLLM >= 0.14.0 |
Please check out the license file for details or contact legal@embedl.com