Yixin Dong
Yixin Dong
Hi @zlwu92, thanks for asking questions about beginning to use XGrammar and testing about llama.cpp. For beginners, I would suggest following [our tutorial](https://xgrammar.mlc.ai/docs/how_to/json_generation.html) that describes how to use xgrammar and...
@YosanHo I think such error should not occur in the latest version of xgrammar. If there are still problems, feel free to show the error message and I can check...
@YosanHo Interesting, I will take a look. Could you share the input, schema and command to launch sglang?
Hi @arunpatala, thanks for the question! The repetition range `{min, max}` is now supported by regex and ebnf, so you can use them to constraint the generation format. Regarding json...
Yes. We will update our doc to cover the supported and unsupported keywords soon.
Hi @gryffindor-rr, thanks for asking that. We are recently working on a project aiming to enhance the completeness of json schema support. We expect almost all features (except some high-level...
@gongwei-130 Thanks for the contribution! Avoiding malformed structural tags is definitely useful. By definition, the structural tag does not allow a null schema as input. Shall we just reject such...
@qeternity Yes the CI fails because xgrammar's new version contains a kernel not compatible with CUDA=12.4. We have fixed this issue (https://github.com/mlc-ai/xgrammar/pull/206) and will release a new version later.
@AlbertoCastelo Thanks for asking that! There is no plan yet to add support for ruby. But feel free to add it if you are interested!