chainlit
chainlit copied to clipboard
Chainlit
Build Conversational AI in minutes ⚡️
**Is your feature request related to a problem? Please describe.** Code pasted by user should be **preserved and rendered** as is **Describe the solution you'd like** Detect pasted code and...
Currently, starters are set before on_chat_start fires, meaning you are locked to a set of starters regardless of what is built within the on_chat_start wrapper. This matters because in the...
Ability to trigger a new chat and clear UI messages with code (particularly within on_chat_resume). This would be very useful to support cases where the user can have an option...
When starting a /command there should be a way to listen to it on the server and provide completions for it. These completions could then be used in the frontend...
**Is your feature request related to a problem? Please describe.** When setting latex=true in `config.toml` it expects mathematical formulas to be encased in `$` signs. All popular LLMs are not...
When creating a new chat dialog, the popup may feel unnecessary (Create new chat? This will clear the current messages and start a new chat.). Is there a way we...
**Is your feature request related to a problem? Please describe.** Right now the files directory (`.files`) is hard-coded into the application and is not configurable. This can be a problem...
**Describe the bug** Now there are hardcoded values in both backend and frontend which break `https` and custom host support when developing application with chainlit. **To Reproduce** Steps to reproduce...
**Is your feature request related to a problem? Please describe.** Yes. While Chainlit offers a clean and extensible UI for LLM-powered apps, it currently lacks native support for generative UI...
**Is your feature request related to a problem? Please describe.** I’m deploying a Chainlit app to Google Cloud Run, which has a hard 32MB request size limit enforced by Google...
