-
I'm using Obsidian on Windows 11. I have been using fabric in WSL2 in ubuntu, but I'm doing a lot of terminal work and copying and pasting which is a bore.
I have configured Mesh-AI Fabric patterns…
-
UCM could allow for some really nice UX if we supported prompts. This comes in several forms:
1. A simple way for a user to confirm a question. This could be "download base into lib.base" or "Looks…
-
### Is your feature request related to a problem? Please describe.
Tags are user-created only (besides security) on our filter panels. Though flexible for the user, this is less ideal for finding inf…
-
I believe an effective way to improve overall stability is to use the previously generated transcription results as the next prompt.
-
Right now the type and contents of each item in a prompt has to be defined before executing the prompt. This makes the prompt a very static thing that can't react to a user's input. It would be nice i…
-
**Is your feature request related to a problem? Please describe.**
The initial implementation of iTerms is not context aware.
**Describe the solution you'd like**
Ideally, users are able …
-
## TL; DR
This RFC proposes separating sample generation and reward model scoring from the original rollout process in PPO, enabling users to more flexibly customize sample generation and create sa…
-
### Check for existing issues
- [X] Completed
### Describe the bug / provide steps to reproduce it
All queries/prompts to LLMs through the inline assistant or in a new context window fails with the…
-
I don't know whether this is a user error or if I should Do Something to fix it:
```shell
Bad URL: failed to parse URL: RelativeUrlWithoutBase: relative URL without a base …
-
Hi,
Seems item midjourney_61 and midjourney_65 are the same. Are there plans to update the 1k set?