Temporary Integration of Other AI Models

Ollama

• Download and install ollama from the official website
• Run OLLAMA_ORIGINS="*" ollama serve (allows cross-origin access and starts ollama)
• Go to the settings page of the plugin, and select openAI for translation service
  • api_key: ollama
  • Custom model: llama2, other models
  • Custom URL: http://localhost:11434/v1/chat/completions
• Related reference documents
  • https://github.com/ollama/ollama/issues/2335

Groq

• Go to the settings page of the plugin, and select openAI for translation service
  • api_key: Obtain api_key
  • Custom model: mixtral-8x7b-32768, llama2-70b-4096
  • Custom URL: https://api.groq.com/openai/v1/chat/completions
  • Rate control: 20/min+25/10min
  • Note
    • It is recommended to use only for mouse hover and input enhancement features## Claude
• Deployment of Claude to ChatGPT using Docker/Cloudflare Claude to ChatGPT
• Go to the settings page of the plugin, and select openAI for the translation service
  • api_key: Obtain the api_key
  • Custom model: claude-instant-1, claude-2
  • Custom URL address: http://localhost:8000/v1/chat/completions
  • See Rate Limits for frequency limitations
  • Notes
    • Be aware of the export IP
    • It is recommended to use only for mouse hover and input enhancement features