Output token limit reached

[gitmeatarru]

What should I do about it???

I tried to write “continue,” but I’m getting this error again…

[gitmeatarru]

up

[aghpub]

I have the same problem. The token limit of subagents is seemingly set to 32k which is clearly not enough for the tasks given to them.

[gitmeatarru]

Although in my case ox-alpha has 1 million output tokens, which seems strange, I don’t think there are more than a million thoughts there.

[filson1]

Long running think processes will hit this limit as well.

[gitmeatarru]

So, we need to adapt the environment to this, otherwise, with so many stars, it doesn’t work as it should with larger models.