-
```shell
/Users/himself65/.local/state/fnm_multishells/1113_1729278934946/bin/npm run build
> www@0.0.0 build
> waku build
vite v5.4.9 building SSR bundle for production...
✓ 1744 modules tra…
-
There are no humans behind any of the chatting on the platform, everything is done through AI NL processing. We should endeavor to find implicit ways to identify that the agents you chat with are _not…
-
# LLM Prompt Testing for choosing icon from set of icons
In this issue we test different prompts for choosing an icon for searched product category.
Testing using GPT3.5
-
### Description
During the reading of AGENTLESS : Demystifying LLM-based Software Engineering Agents, I found that the icons were overly enlarged, taking up the entire screen, resulting in simply unr…
-
### Summary
The vscode extension now allows sign in with Stencila for cloud features including usage of LLM credits for prompts in the document. It mainly works on the "happy-path" but the experie…
-
### Feature Request
Hello.
About UX... again... :)
As of GPT 2.7.1, on the main /page:
- the font for the Name of the current LLM is/looks the same as that of other text
- the color of all…
-
I've been trying to use DSPy in different contexts where I see fit, but I've been unsuccessful in obtaining any good results. I have a very long prompt for a classification task that needs to describe…
-
**, the main purpose is …
-
### 🥰 Feature Description
As mentioned in https://github.com/lobehub/lobe-chat/issues/1100#issue-2089755728 (inactive)
I think this would be a great addition, the emojis are very limiting and make…