Jonas Scholz

Hi! I know this issue is quite old, but I am trying the same right now. I want to count bytes when using https, where do I set a new...

I'm having the same issue, for some reason a delay after `onMount` fixes this issue for me. Not sure why:D

Same here with plain JavaScript. Trying to remove the widget as @TomPradat said, getting the timeout after 15-30 Seconds :(

Are there any plans to implement/merge this? I'm wondering if I should wait for it or just implement it myself :D

Is there anything that is blocking this? If not I could give it a try

Having the same issue for [menlo/jan-nano](https://huggingface.co/Menlo/Jan-nano-128k) (obviously, since its just fine-tuned qwen3). Passing in the rope-scaling arg as JSON breaks it. ```bash vllm serve Menlo/Jan-nano-128k \ --host 0.0.0.0 \ --port...

I have no clue about python or vllm, wouldn't a `"rope_scaling": json.loads(os.getenv("ROPE_SCALING", "null")),` be enough? Looks like the vllm engine [always expects a dict](https://github.com/vllm-project/vllm/blob/main/vllm/engine/arg_utils.py#L354C5-L354C17)? Just added this to my [fork](https://github.com/Code42Cate/worker-vllm/commit/c923536844875568982e680f25b4f652d05ff547),...

Ah yeah I was too lazy to add the other ones, yours is better :D

Fork the repository with the fixed code (mine or preferably @FuxMak) and then deploy from your forked repository instead of from their provided Docker image

My integration tests don’t really clean up after themselves, but it seems like the existing ones don’t either. Was that intentional, or just something that never got done? If it...