Text Generation
Transformers
Safetensors
qwen3
merged
lora
chinese
novel
conversational
text-generation-inference
Instructions to use wonzer/qwen3_lora_mix with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use wonzer/qwen3_lora_mix with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="wonzer/qwen3_lora_mix") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("wonzer/qwen3_lora_mix") model = AutoModelForCausalLM.from_pretrained("wonzer/qwen3_lora_mix", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use wonzer/qwen3_lora_mix with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "wonzer/qwen3_lora_mix" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "wonzer/qwen3_lora_mix", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/wonzer/qwen3_lora_mix
- SGLang
How to use wonzer/qwen3_lora_mix with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "wonzer/qwen3_lora_mix" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "wonzer/qwen3_lora_mix", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "wonzer/qwen3_lora_mix" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "wonzer/qwen3_lora_mix", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use wonzer/qwen3_lora_mix with Docker Model Runner:
docker model run hf.co/wonzer/qwen3_lora_mix
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,67 +1,67 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
base_model: Qwen/Qwen3-4B
|
| 4 |
-
tags:
|
| 5 |
-
- merged
|
| 6 |
-
- lora
|
| 7 |
-
- chinese
|
| 8 |
-
- novel
|
| 9 |
-
- text-generation
|
| 10 |
-
library_name: transformers
|
| 11 |
-
pipeline_tag: text-generation
|
| 12 |
-
---
|
| 13 |
-
|
| 14 |
-
# Qwen Novel Generation Model (LoRA Merged)
|
| 15 |
-
|
| 16 |
-
这是一个基于 Qwen/Qwen3-4B 和自定义LoRA适配器合并的模型,专门用于中文小说生成。
|
| 17 |
-
|
| 18 |
-
## 模型详情
|
| 19 |
-
|
| 20 |
-
- **基础模型**: Qwen/Qwen3-4B
|
| 21 |
-
- **LoRA路径**: D:/Work/Project/lora_novel/lora_output/final
|
| 22 |
-
- **合并方法**: PeftModel.merge_and_unload()
|
| 23 |
-
- **用途**: 中文
|
| 24 |
-
- **语言**: 中文
|
| 25 |
-
|
| 26 |
-
## 使用方法
|
| 27 |
-
|
| 28 |
-
```python
|
| 29 |
-
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 30 |
-
|
| 31 |
-
# 加载模型和tokenizer
|
| 32 |
-
model = AutoModelForCausalLM.from_pretrained("your-username/model-name")
|
| 33 |
-
tokenizer = AutoTokenizer.from_pretrained("your-username/model-name")
|
| 34 |
-
|
| 35 |
-
# 生成文本
|
| 36 |
-
def generate_story(prompt, max_length=500):
|
| 37 |
-
inputs = tokenizer(prompt, return_tensors="pt")
|
| 38 |
-
outputs = model.generate(
|
| 39 |
-
**inputs,
|
| 40 |
-
max_length=max_length,
|
| 41 |
-
temperature=0.7,
|
| 42 |
-
do_sample=True,
|
| 43 |
-
top_p=0.9,
|
| 44 |
-
repetition_penalty=1.1
|
| 45 |
-
)
|
| 46 |
-
return tokenizer.decode(outputs[0], skip_special_tokens=True)
|
| 47 |
-
|
| 48 |
-
# 示例
|
| 49 |
-
story = generate_story("请写一个关于友情的故事:")
|
| 50 |
-
print(story)
|
| 51 |
-
```
|
| 52 |
-
|
| 53 |
-
## 生成示例
|
| 54 |
-
|
| 55 |
-
**输入**: "请写一个关于友情的故事:"
|
| 56 |
-
|
| 57 |
-
**输出**: [模型会生成相应的故事内容]
|
| 58 |
-
|
| 59 |
-
## 注意事项
|
| 60 |
-
|
| 61 |
-
- 这是一个合并后的完整模型,包含了LoRA的所有改进
|
| 62 |
-
- 适合用于中文创作场景
|
| 63 |
-
- 建议使用适当的生成参数以获得最佳效果
|
| 64 |
-
|
| 65 |
-
## 许可证
|
| 66 |
-
|
| 67 |
-
本模型遵循Apache 2.0许可证。
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: Qwen/Qwen3-4B
|
| 4 |
+
tags:
|
| 5 |
+
- merged
|
| 6 |
+
- lora
|
| 7 |
+
- chinese
|
| 8 |
+
- novel
|
| 9 |
+
- text-generation
|
| 10 |
+
library_name: transformers
|
| 11 |
+
pipeline_tag: text-generation
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# Qwen Novel Generation Model (LoRA Merged)
|
| 15 |
+
|
| 16 |
+
这是一个基于 Qwen/Qwen3-4B 和自定义LoRA适配器合并的模型,专门用于中文小说生成。
|
| 17 |
+
|
| 18 |
+
## 模型详情
|
| 19 |
+
|
| 20 |
+
- **基础模型**: Qwen/Qwen3-4B
|
| 21 |
+
- **LoRA路径**: D:/Work/Project/lora_novel/lora_output/final
|
| 22 |
+
- **合并方法**: PeftModel.merge_and_unload()
|
| 23 |
+
- **用途**: 中文文本生成
|
| 24 |
+
- **语言**: 中文
|
| 25 |
+
|
| 26 |
+
## 使用方法
|
| 27 |
+
|
| 28 |
+
```python
|
| 29 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 30 |
+
|
| 31 |
+
# 加载模型和tokenizer
|
| 32 |
+
model = AutoModelForCausalLM.from_pretrained("your-username/model-name")
|
| 33 |
+
tokenizer = AutoTokenizer.from_pretrained("your-username/model-name")
|
| 34 |
+
|
| 35 |
+
# 生成文本
|
| 36 |
+
def generate_story(prompt, max_length=500):
|
| 37 |
+
inputs = tokenizer(prompt, return_tensors="pt")
|
| 38 |
+
outputs = model.generate(
|
| 39 |
+
**inputs,
|
| 40 |
+
max_length=max_length,
|
| 41 |
+
temperature=0.7,
|
| 42 |
+
do_sample=True,
|
| 43 |
+
top_p=0.9,
|
| 44 |
+
repetition_penalty=1.1
|
| 45 |
+
)
|
| 46 |
+
return tokenizer.decode(outputs[0], skip_special_tokens=True)
|
| 47 |
+
|
| 48 |
+
# 示例
|
| 49 |
+
story = generate_story("请写一个关于友情的故事:")
|
| 50 |
+
print(story)
|
| 51 |
+
```
|
| 52 |
+
|
| 53 |
+
## 生成示例
|
| 54 |
+
|
| 55 |
+
**输入**: "请写一个关于友情的故事:"
|
| 56 |
+
|
| 57 |
+
**输出**: [模型会生成相应的故事内容]
|
| 58 |
+
|
| 59 |
+
## 注意事项
|
| 60 |
+
|
| 61 |
+
- 这是一个合并后的完整模型,包含了LoRA的所有改进
|
| 62 |
+
- 适合用于中文创作场景
|
| 63 |
+
- 建议使用适当的生成参数以获得最佳效果
|
| 64 |
+
|
| 65 |
+
## 许可证
|
| 66 |
+
|
| 67 |
+
本模型遵循Apache 2.0许可证。
|