Jesse Luoto

Zeromq not needed anymore since v2 was released, should work now on Windows. I think I could add some e2e test for running on Windows, as [GH runners have a...

Sounds like there has been some change in OpenAI API, since previously there was no issue. I'll check what the rate limits are currently, should be easy to fix

> Rate limits are imposed at the [organization level](https://platform.openai.com/docs/guides/production-best-practices), not user level. https://platform.openai.com/docs/guides/rate-limits?context=tier-free I was first thinking about throttling the requests to an allowed level, but I think it makes...

Yes, great point! I think it would make sense to even parametrize the model as a command line argument 🤔

That's a great addition, thanks! I was searching for similar tools like webcrack, but I think I've missed `wakaru`. I'll definitely give it a try!

It seems that `wakaru` does not yet have a (functioning) npm package, which makes usage harder. I'll add an issue to their repo about it. I think I can come...

Fixed at v2, no Python needed anymore

I actually dropped guidance (and python) in v2, I'm now using [gbnf grammars](https://github.com/ggerganov/llama.cpp/blob/master/grammars/README.md) to force a specific output instead of using Guidance.

Not possible at the moment, but should be straightforward to implement if you'd like to give it a shot You can check `LlamaCpp` docs from Guidance and change (preferably parametrize)...

There's now v2 that runs on top of llama.cpp, so adding llama3 support should be even more straightforward. @0xrsydn which version of llama3 were you planning to run? I could...