Update README and fix serve CLI import

Reorganize Quick Start into clear sections (chat, serve, Docker).
Move Qwen3.5 models to top of model list. Fix serve CLI to use
vllm's FlexibleArgumentParser and async run_server.

Made-with: Cursor

Author: zhijian-liu

README.md

@@ -19,28 +19,50 @@ State-of-the-art INT4 quantization for LLMs. ParoQuant uses learned pairwise rot
 
 ## Quick Start
 
-**NVIDIA GPU:**
+### Interactive Chat
 
 ```bash
+# NVIDIA GPU
 pip install "paroquant[vllm]"
 python -m paroquant.cli.chat --model z-lab/Qwen3-8B-PARO
 
-# or with Docker
-docker run --pull=always --rm -it --gpus all --ipc=host \
-  ghcr.io/z-lab/paroquant:chat --model z-lab/Qwen3-8B-PARO
+# Apple Silicon
+pip install "paroquant[mlx]"
+python -m paroquant.cli.chat --model z-lab/Qwen3-8B-PARO
 ```
 
-**Apple Silicon:**
+### OpenAI-Compatible API Server
 
 ```bash
-pip install "paroquant[mlx]"
-python -m paroquant.cli.chat --model z-lab/Qwen3-8B-PARO
+pip install "paroquant[vllm]"
+python -m paroquant.cli.serve --model z-lab/Qwen3-8B-PARO
+```
+
+### Docker
+
+```bash
+# Interactive chat
+docker run --pull=always --rm -it --gpus all --ipc=host \
+  ghcr.io/z-lab/paroquant:chat --model z-lab/Qwen3-8B-PARO
+
+# API server (port 8000)
+docker run --pull=always --rm -it --gpus all --ipc=host -p 8000:8000 \
+  ghcr.io/z-lab/paroquant:serve --model z-lab/Qwen3-8B-PARO
 ```
 
 ## Models
 
 All models are available on [Hugging Face](https://huggingface.co/collections/z-lab/paroquant). Swap the model name in the commands above to try any of them.
 
+**Qwen3.5**
+
+| Model | Checkpoint |
+|---|---|
+| Qwen3.5-0.8B | [`z-lab/Qwen3.5-0.8B-PARO`](https://huggingface.co/z-lab/Qwen3.5-0.8B-PARO) |
+| Qwen3.5-2B | [`z-lab/Qwen3.5-2B-PARO`](https://huggingface.co/z-lab/Qwen3.5-2B-PARO) |
+| Qwen3.5-4B | [`z-lab/Qwen3.5-4B-PARO`](https://huggingface.co/z-lab/Qwen3.5-4B-PARO) |
+| Qwen3.5-9B | [`z-lab/Qwen3.5-9B-PARO`](https://huggingface.co/z-lab/Qwen3.5-9B-PARO) |
+
 **Qwen3**
 
 | Model | Checkpoint |
@@ -51,15 +73,6 @@ All models are available on [Hugging Face](https://huggingface.co/collections/z-
 | Qwen3-8B | [`z-lab/Qwen3-8B-PARO`](https://huggingface.co/z-lab/Qwen3-8B-PARO) |
 | Qwen3-14B | [`z-lab/Qwen3-14B-PARO`](https://huggingface.co/z-lab/Qwen3-14B-PARO) |
 
-**Qwen3.5**
-
-| Model | Checkpoint |
-|---|---|
-| Qwen3.5-0.8B | [`z-lab/Qwen3.5-0.8B-PARO`](https://huggingface.co/z-lab/Qwen3.5-0.8B-PARO) |
-| Qwen3.5-2B | [`z-lab/Qwen3.5-2B-PARO`](https://huggingface.co/z-lab/Qwen3.5-2B-PARO) |
-| Qwen3.5-4B | [`z-lab/Qwen3.5-4B-PARO`](https://huggingface.co/z-lab/Qwen3.5-4B-PARO) |
-| Qwen3.5-9B | [`z-lab/Qwen3.5-9B-PARO`](https://huggingface.co/z-lab/Qwen3.5-9B-PARO) |
-
 **Llama**
 
 | Model | Checkpoint |
@@ -106,10 +119,10 @@ python -m paroquant.cli.convert \
 
 | Image | Purpose |
 |---|---|
-| `ghcr.io/z-lab/paroquant:latest` | Optimization & evaluation |
-| `ghcr.io/z-lab/paroquant:chat` | Interactive chat (CUDA 13.0) |
-| `ghcr.io/z-lab/paroquant:serve` | OpenAI-compatible API server |
+| `ghcr.io/z-lab/paroquant:chat` | Interactive chat |
 | `ghcr.io/z-lab/paroquant:chat-cu129` | Interactive chat (CUDA 12.9) |
+| `ghcr.io/z-lab/paroquant:serve` | OpenAI-compatible API server |
+| `ghcr.io/z-lab/paroquant:latest` | Optimization & evaluation |
 | `ghcr.io/z-lab/paroquant:eval` | Reasoning task evaluation |
 
 ## Citation

paroquant/cli/serve.py

@@ -1,17 +1,16 @@
-"""Thin wrapper around ``vllm serve`` that auto-registers the ParoQuant quantization plugin."""
-
 from __future__ import annotations
 
 import asyncio
 import sys
 
-import paroquant.inference.backends.vllm.plugin  # noqa: F401 — registers quantization config
 from vllm.entrypoints.openai.api_server import (
     FlexibleArgumentParser,
     make_arg_parser,
     run_server,
 )
 
+import paroquant.inference.backends.vllm.plugin  # noqa: F401 — registers quantization config
+
 
 def main():
     parser = make_arg_parser(FlexibleArgumentParser())