README.md

---
base_model: unsloth/Qwen2.5-Coder-32B-Instruct-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2
- trl
license: apache-2.0
language:
- en
---

# Qwen2.5-Coder-32B-Instruct-WMX
Pre-fine-tuned LoRA adapters for unsloth/Qwen2.5-Coder-32B-Instruct.

**This lora adapters have been fine-tuned for WMX services using the folowing datasets.**
- https://huggingface.co/datasets/Jake5/movensys-info
- https://huggingface.co/datasets/Jake5/wmx-doc-user
- https://huggingface.co/datasets/Jake5/wmx-doc-robot

## Version v0.9
- Source: lora_model
- Base model: unsloth/Qwen2.5-Coder-32B-Instruct
- Uploaded on: 2025-09-12

## Usage
```python
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
base_model = AutoModelForCausalLM.from_pretrained("unsloth/Qwen2.5-Coder-32B-Instruct")
model = PeftModel.from_pretrained(base_model, "Jake5/Qwen2.5-Coder-32B-Instruct-WMX", subfolder="adapters_v0.9")
tokenizer = AutoTokenizer.from_pretrained("Jake5/Qwen2.5-Coder-32B-Instruct-WMX", subfolder="adapters_v0.9")
```

## vLLM Serving
```bash
python -m vllm.entrypoints.openai.api_server \
    --model unsloth/Qwen2.5-Coder-32B-Instruct \
    --lora-modules my-lora=Jake5/Qwen2.5-Coder-32B-Instruct-WMX/adapters_v0.9 \
    --dtype bfloat16 \
    --port 8000
```