BrickJZ commited on
Commit
b67dc8a
·
verified ·
1 Parent(s): 0805330

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +51 -3
README.md CHANGED
@@ -1,3 +1,51 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ language:
4
+ - zh
5
+ - en
6
+ tags:
7
+ - llama.cpp
8
+ - TQ3
9
+ - quantization
10
+ - Windows
11
+ - NVIDIA
12
+ - GGUF
13
+ ---
14
+
15
+ # llama.cpp-TQ3 专用推理环境 (Windows/NVIDIA版)
16
+ # llama.cpp-TQ3 Inference Environment (Windows/NVIDIA Edition)
17
+
18
+ ## 简介 | Intro
19
+ 这是一个预编译的 `llama.cpp` 环境,专为 **TQ3 量化模型**设计,支持 NVIDIA 显卡在 Windows 上一键运行。
20
+ A pre-built `llama.cpp` environment optimized for **TQ3 quantized models**, enabling one-click inference on NVIDIA GPUs for Windows users.
21
+
22
+ ## 核心特性 | Key Features
23
+ ✅ 原生支持 TQ3 格式(普通 llama.cpp 无法运行)
24
+ ✅ 已编译 CUDA 加速,专为 NVIDIA 显卡优化
25
+ ✅ 免配置依赖,解压即用,不包含模型权重
26
+ ✅ 支持命令行与 Web 服务两种运行方式
27
+
28
+ ✅ Native TQ3 support (works with models standard llama.cpp cannot run)
29
+ ✅ CUDA-accelerated, optimized for NVIDIA GPUs
30
+ ✅ No dependencies required — just extract and run (model weights not included)
31
+ ✅ Supports both CLI and Web server modes
32
+
33
+ ## 使用方法 | Usage
34
+ 1. **下载解压**:将文件解压到纯英文路径
35
+ 2. **放入模型**:把 `.tq3.gguf` 格式的模型放到目录下
36
+ 3. **启动运行**:使用 `llama-cli.exe` 或 `llama-server.exe` 加载模型
37
+
38
+ 1. **Download & Extract**: Unzip to a folder with an English-only path
39
+ 2. **Add Model**: Place your `.tq3.gguf` model in the same directory
40
+ 3. **Run**: Use `llama-cli.exe` or `llama-server.exe` to start inference
41
+
42
+ ## 注意事项 | Notes
43
+ - 仅支持 **NVIDIA 显卡**,AMD 显卡暂不兼容
44
+ - 本项目不包含任何模型文件,请自行获取并遵守对应开源协议
45
+
46
+ - **NVIDIA-only**: Not compatible with AMD GPUs
47
+ - This package does not include model weights. Please obtain them legally and comply with their licenses.
48
+
49
+ ## 致谢 | Credits
50
+ - 核心源码: turbo-tan/llama.cpp-tq3
51
+ - TQ3 量化: YTan2000