Clouditera / Clouditera/Clouditera.github.io
基于GGUF格式的云起无垠SecGPT大语言模型
- Dominant language
- HTML
- Stars
- 628
- Forks
- 46
- PR merge metrics
- No merged PRs in 30d
Description
# GGUF模型链接
https://huggingface.co/LingJingMaster/Clouditera-SecGPT-GGUF/blob/main/README.md?code=true
## SecGPT 网络安全大模型
### **项目**
- [GitHub](https://github.com/Clouditera/SecGPT)
- [原版Pytorch模型](https://huggingface.co/clouditera/secgpt)
### **简介**
- 随着大语言模型的崛起,网安大模型也掀起了一股热潮,本人在逛 GitHub 时偶然发现了云起无垠开源的 SecGPT,但官方调用脚本中使用了 Cuda,且没有提供 GGUF 版本,故使用了 [llama.cpp](https://github.com/ggerganov/llama.cpp) 的 convert 脚本进行转换,并上传至huggingface
### **测试设备**
- MacBook Pro 16 寸
- M3 Max
- 48 GB
### **Usage**
- 分为 `secgpt1.5.gguf` , `secgpt.gguf` 与 `secgpt-mini.gguf` 三个版本
- `secgpt1.5.gguf`基于qwen2, 需 28.5 G 显存
- `secgpt.gguf`基于baichuan, 需 26.5 G 显存
- `secgpt-mini.gguf`基于qwen2, 需 1.6 G 显存
- 使用方法
- 将 GGUF 导入[LM Studio](https://lmstudio.ai/),并使用 `secgpt-all.json` 作为参数配置
Contributor guide
No contributing guide indexed for this repository
Research direction
The payload names no repository file, test, or entry point. Start by reviewing the linked Hugging Face README and the `secgpt-all.json` usage configuration; completion cannot be verified until the repository’s intended location and acceptance criteria for this model information are identified.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, huggingface
- Domain
- ai, content, security
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100