-
Notifications
You must be signed in to change notification settings - Fork 1.9k
Description
Describe the bug
When initially setting up Letta server from Letta Cloud, it's not possible to configure and change the default http://localhost:8283 server. Which, if you want to run that on another machine or server in your network is weird. For that to work I actually had to add a route on the localhost machine, for localhost:8283 to point to another address. It would have made more sense if we could configure the localhost:8283 address to another ip address. Or to at least be able to delete or disable that self-hosted server. It's not possible to edit the configuration now.
Please describe your setup
-
How are you running Letta?
- Letta Desktop Windows x64 v0.11.5
-
Describe your setup
- Windows 11 x64
- Seperate Debian machine with Letta Server installed in a Docker
Screenshots


Additional context
Add any other context about the problem here.
- What model you are using
Agent File (optional)
Please attach your .af
file, as this helps with reproducing issues.
If you're not using OpenAI, please provide additional information on your local LLM setup:
Local LLM details
If you are trying to run Letta with local LLMs, please provide the following information:
- The exact model you're trying to use (e.g.
dolphin-2.1-mistral-7b.Q6_K.gguf
) - The local LLM backend you are using (web UI? LM Studio?)
- Your hardware for the local LLM backend (local computer? operating system? remote RunPod?)