[Bug] mlc_llm chat throws errors for model mlc-ai/Qwen1.5-1.8B-Chat-q4f16_1-MLC

[BodhiHu]

🐛 Bug

Hello,

HF://mlc-ai/Qwen1.5-1.8B-Chat-q4f16_1-MLC seems to be incomplete:

missing max_batch_size in mlc-chat-config.json;
no tokenizers found under Qwen1.5-1.8B-Chat-q4f16_1-MLC/

And above two missings will cause mlc_llm chat ... throw errors.

To Reproduce

Steps to reproduce the behavior:

1. mlc_llm chat mlc-ai/Qwen1.5-1.8B-Chat-q4f16_1-MLC

Expected behavior

Environment

• Platform (e.g. WebGPU/Vulkan/IOS/Android/CUDA): Mac M1, Metal
• Operating system (e.g. Ubuntu/Windows/MacOS/...): MacOS
• Device (e.g. iPhone 12 Pro, PC+RTX 3090, ...)
• How you installed MLC-LLM (conda, source): yes
• How you installed TVM-Unity (pip, source):
• Python version (e.g. 3.10):
• GPU driver version (if applicable):
• CUDA/cuDNN version (if applicable):
• TVM Unity Hash Tag (python -c "import tvm; print('\n'.join(f'{k}: {v}' for k, v in tvm.support.libinfo().items()))", applicable if you compile models):
• Any other relevant information:

Additional context

[mengshyu]

Hi @BodhiHu, I've updated the config file, can you try it again with latest mlc llm, thanks.

https://huggingface.co/mlc-ai/Qwen1.5-1.8B-Chat-q4f16_1-MLC/commit/09f17e66c5fa19bb938898e23d7997ac06371abf