Luca Beurer-Kellner

Thanks for the suggestion. We will try to add some more information to the docs. In general, simple constraints like len(TOKEN(..))), STOPS_AT and STOPS_BEFORE are possible with ChatGPT. More advanced...

Thanks for linking, we have indeed already looked at this paper and investigated its applicability. The graph in 3.2 is impressive, but unfortunately we could not yet reproduce this result...

I just pushed a fix to `main` which resolves it. Thanks for reporting it, `@dataclass` support is still in Preview, so this kind of feedback is very helpful.

Thanks for the PR, conceptually the changes look good to me. So the logs we are seeing, suggest that the inference server somehow switches between both model continuously, even though...

We definitely want to support this. They have already been some discussion on Discord surrounding this. See the #dev channel for more details.

Thanks for the reproduction package. We will have a look.

It would be interesting to see an implementation of ToT in LMQL. I am not sure about immediate token savings, as ToT is mostly compositional. Either way having it as...

Adding a traditional agent example will definitely be interesting to have in the documentation. For now, you can already have a look at the Wikipedia example (https://lmql.ai/#wiki), which showcases tool...

I will keep this issue around, until we add a more thorough agent example to the documentation.