gandalf/secubox-openwrt
1
1
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
b8fb58b392
secubox-openwrt/package/secubox/secubox-app-ollama/README.zh.md
CyberMind-FR ccfb58124c docs: Add trilingual documentation (French and Chinese translations) 
Add complete French (fr) and Chinese (zh) translations for all documentation:

- Root files: README, CHANGELOG, SECURITY, BETA-RELEASE
- docs/: All 16 core documentation files
- DOCS/: All 19 deep-dive documents including embedded/ and archive/
- package/secubox/: All 123+ package READMEs
- Misc: secubox-tools/, scripts/, EXAMPLES/, config-backups/, streamlit-apps/

Total: 346 translation files created

Each file includes language switcher links for easy navigation between
English, French, and Chinese versions.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2026-03-20 10:00:18 +01:00

1.3 KiB
Raw Blame History

English | Francais | 中文

SecuBox Ollama - 本地 LLM 运行时

在您的 OpenWrt 设备上本地运行大型语言模型。提供兼容 OpenAI 的 REST API，原生支持 ARM64。支持 LLaMA、Mistral、Phi、Gemma 和其他开放模型。

安装

opkg install secubox-app-ollama

配置

UCI 配置文件：/etc/config/ollama

uci set ollama.main.enabled='1'
uci set ollama.main.bind='0.0.0.0'
uci set ollama.main.port='11434'
uci set ollama.main.model_dir='/srv/ollama/models'
uci commit ollama

使用方法

ollamactl start              # 启动 Ollama 服务
ollamactl stop               # 停止 Ollama 服务
ollamactl status             # 显示服务状态
ollamactl pull <model>       # 下载模型
ollamactl list               # 列出已安装的模型
ollamactl remove <model>     # 删除模型
ollamactl run <model>        # 运行交互式聊天

API

兼容 OpenAI 的端点位于 http://<host>:11434：

curl http://localhost:11434/api/generate -d '{
  "model": "llama3.2",
  "prompt": "Hello"
}'

支持的模型

LLaMA 3.x、Mistral、Phi-3、Gemma 2、CodeLlama，以及任何兼容 GGUF 的模型。

依赖

  • jsonfilter
  • wget-ssl

许可证

Apache-2.0