- feat: chat with files (including text files and images)
- feat: set selected prompt as system prompt
- fix markdown rendering issue
- bump dependencies for security fixes
- Add a set of new models from OpenAI, Gemini and Claude, including o1, claude 3.5 and gemini-2.0-flash-thinking models.
- Bump default model to gpt-4o from gpt-3.5-turbo.
- Add support of prompt manager and chat with your own prompts (use # to search)
- Fix the version compatibility issue
- Add support of Google Generative AI models
- Refine the project and reduce extension size (web searching is temporally removed and will be added back in the future)
- Add support for Anthropic Claude 3.5
- Support GPT-4o
- Support custom model names for local or self-hosted LLMs
- Add support of Serper and Bing search
- Add searching support for Claude models
- Add support for Anthropic Claude 3
- Add support for Google Custom Search
- Set "chatgpt.gpt3.googleCSEApiKey" and "chatgpt.gpt3.googleCSEId" to enable this feature
- Add support of customized baseURL
- Bump depdencies and fix CVEs
- Add support of latest OpenAI models (e.g. GPT-4 Turbo and so on)
- Add support of reading OpenAI Key from environment variable "OPENAI_API_KEY" when it is not set in vscode configure file
- Add a few samples for typical configurations
- Switch to a new LOGO
- Added streaming output support
- Bump dependencies to fix the potential security issues
- Bump dependencies to fix the potential security issues
- Bump dependencies to fix the potential security issues
- Fix the default base URL for OpenAI
- Cleanup the unused parameters and functions
- Fix the conversation stuck issues