Instructions to use lemonbucket/yuwen-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lemonbucket/yuwen-v1 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="lemonbucket/yuwen-v1",
	filename="yuwen-v1-f16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use lemonbucket/yuwen-v1 with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf lemonbucket/yuwen-v1:F16
# Run inference directly in the terminal:
llama cli -hf lemonbucket/yuwen-v1:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf lemonbucket/yuwen-v1:F16
# Run inference directly in the terminal:
llama cli -hf lemonbucket/yuwen-v1:F16

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf lemonbucket/yuwen-v1:F16
# Run inference directly in the terminal:
./llama-cli -hf lemonbucket/yuwen-v1:F16

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf lemonbucket/yuwen-v1:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf lemonbucket/yuwen-v1:F16

Use Docker

docker model run hf.co/lemonbucket/yuwen-v1:F16

LM Studio
Jan

vLLM

How to use lemonbucket/yuwen-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lemonbucket/yuwen-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lemonbucket/yuwen-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lemonbucket/yuwen-v1:F16

Ollama
How to use lemonbucket/yuwen-v1 with Ollama:
```
ollama run hf.co/lemonbucket/yuwen-v1:F16
```

Unsloth Studio

How to use lemonbucket/yuwen-v1 with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for lemonbucket/yuwen-v1 to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for lemonbucket/yuwen-v1 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for lemonbucket/yuwen-v1 to start chatting

How to use lemonbucket/yuwen-v1 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf lemonbucket/yuwen-v1:F16

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "lemonbucket/yuwen-v1:F16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use lemonbucket/yuwen-v1 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf lemonbucket/yuwen-v1:F16

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default lemonbucket/yuwen-v1:F16

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use lemonbucket/yuwen-v1 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf lemonbucket/yuwen-v1:F16

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "lemonbucket/yuwen-v1:F16" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use lemonbucket/yuwen-v1 with Docker Model Runner:
```
docker model run hf.co/lemonbucket/yuwen-v1:F16
```

Lemonade

How to use lemonbucket/yuwen-v1 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull lemonbucket/yuwen-v1:F16

Run and chat with the model

lemonade run user.yuwen-v1-F16

List all available models

lemonade list

yuwen-v1: 语文阅读理解微调模型 (GGUF)

模型简介

yuwen-v1 是柠檬桶 (lemonbucket) 的第一个语文微调模型，基于 Qwen3-0.6B 进行 LoRA 微调。

本仓库仅包含 GGUF 格式文件，适合通过 Ollama / llama.cpp 部署，不支持 Transformers 直接加载。

基座模型

项目	说明
模型	Qwen/Qwen3-0.6B
参数量	~0.6B
微调方式	LoRA (PEFT)

仓库文件

文件	说明	大小
`yuwen-v1-f16.gguf`	F16 半精度 GGUF	~1.1 GB
`Modelfile`	Ollama 部署配置	-

快速开始

Ollama 部署（推荐）

# 1. 下载本仓库的 GGUF 文件和 Modelfile
# 2. 创建 Ollama 模型
ollama create yuwen-v1 -f Modelfile

# 3. 运行
ollama run yuwen-v1

# 或直接通过 API 调用
curl http://localhost:11434/api/generate -d '{
  "model": "yuwen-v1",
  "prompt": "请分析《荷塘月色》中通感手法的运用。",
  "stream": false
}'

llama.cpp

# llama.cpp 基本用法
./llama-cli   -m yuwen-v1-f16.gguf   --temp 0.3 --top-k 20 --top-p 0.85   -p "请分析《荷塘月色》中通感手法的运用。"

系统提示词

你是 yuwen-v1，一个由柠檬桶（LemonBucket）基于 Qwen3-0.6B 通过 LoRA 微调训练的语文知识 AI 助手。你专注于语文知识的解答，包括病句、文章分析、环境描写、叙述顺序、抒情方式等。

评测结果

本系列模型在 2026 年高考语文现代文阅读测试中进行了 9 模型横向评测。

项目	说明
测试标准	2026 年全国 I 卷语文现代文阅读，满分 39 分（选择题 15 分 + 简答题 24 分）
阅卷评判官	GLM-5.2（6 维度评分：答题规范性、长文本处理、基础理解、常识储备、信息准确性、深度分析，满分 60 分）
完整报告	lemonbucket/yuwen-v3 仓库的 `yuwen测评/` 目录

关于思考模式 (Thinking Mode)

重要: 由于微调时训练数据中未包含思维链 (<think>) 相关内容，模型的思维链输出存在异常。强烈推荐关闭思考模式使用。

部署方式关闭方法

Ollama Modelfile 已预配置，无需额外操作

Transformers V3 仓库 chat_template 已修改为默认关闭

其他框架 请勿传入 enable_thinking=True 参数

若上述配置失效（框架升级、模板被覆盖等），请手动修改 Modelfile、chat_template.jinja 或 tokenizer_config.json 中的 chat_template 字段，移除 <think> 相关内容即可。

部署方式	关闭方法
Ollama	Modelfile 已预配置，无需额外操作
Transformers	V3 仓库 `chat_template` 已修改为默认关闭
其他框架	请勿传入 `enable_thinking=True` 参数

微调方向

本系列模型专门针对语文现代文阅读理解进行微调，涵盖：

信息类文本阅读（选择题 + 简答题）
文学类文本阅读（选择题 + 简答题 + 续写）
答题规范：分点作答、术语规范、踩分精准

许可证

本模型基于 Apache 2.0 协议开源。

项目	许可
LoRA 适配器权重	Apache 2.0（继承自基座模型）
合并后完整模型权重	Apache 2.0（继承自基座模型）
基座模型 Qwen3-0.6B / Qwen3.5-0.8B	Apache 2.0（由 Alibaba Qwen Team 开源）

使用时需同时遵守 Apache 2.0 协议及 Qwen 模型许可条款。

关于作者

我是刚接触模型微调的新手，初次使用 Hugging Face 平台。如果仓库存在不规范之处（文件缺失、配置有误、描述不清等），还请通过邮件反馈：575244421@qq.com，不胜感激！

致谢

基座模型: Qwen 系列 by Alibaba Qwen Team
微调框架: HuggingFace Transformers + PEFT
训练者: 柠檬桶 (lemonbucket)

Downloads last month: 30

GGUF

Model size

0.6B params

Architecture

qwen3

Hardware compatibility

16-bit

Model tree for lemonbucket/yuwen-v1

Base model

Qwen/Qwen3-0.6B-Base

Finetuned

Qwen/Qwen3-0.6B

Quantized

(376)

this model