Arsham Shirvani
Arsham Shirvani
@PilotDevo have you noticed this happening when context usage is high? I have seen this happening when the context is over 150k out of 200k. A simple test is to...
Thanks rekram, however I was referring to the last few messages and not the beginning.
Was there a tool call just before that? If that was the case, the result of the tool call must have been large. Your only option is to run `/undo`...
I have good results with these instructions, hope this helps you too: ```markdown **Behaviours**: - No praise, fluff, apologies, or hedging - Challenge user premises when flawed - Disagree when...
I have experienced this too. For now you could do this: type `/undo`, then enter. It will resend your last prompt and it usually works. Sometimes I had to restart...
This is actually handy. I was wondering when we would see such thing. Basically during the reasoning phase, the model could use a tool call to enrich its reasoning.
It is lovely. A small problem:
This is an interesting idea. I would recommend keeping the current `!` behaviour, and only use a `!!` for entering this mode .
I have an implementation using the `experimental.chat.messages.transform` event. I am still working on it and will opensource it at some point. However, I think if the community would come up...
I absolutely love this! I know this is not the scope of this PR, but I was wondering if we could slip in optional model to the task tool. Ignore...