Chat with Llama 3.2 1B

[shubhamgupto]

Hello,

I want to run Llama 3.2 1B on my jetson orin nano using nanollm interface. I have been granted access to the llama 3.2 models but im not sure how to share my hf token with the container.

HUGGINGFACE_KEY=<>	 \
MLC_VERSION=0.1.2 \
jetson-containers run $(autotag nano_llm) \
  python3 -m nano_llm.chat --api=mlc \
    --model meta-llama/Llama-3.2-1B

let me know if there's a better way to do this, thanks

[shubhamgupto]

I was able to first run the container and do huggingface-cli login and now i am able to download the weights, would be nice if its a 1 command like rest of the models

[dusty-nv]

Hi @shubhamgupto , you can set it by setting HUGGINGFACE_TOKEN like shown here: https://www.jetson-ai-lab.com/tutorial_nano-llm.html#containers

[shubhamgupto]

Hey @dusty-nv this is great, thanks, any idea why the following error occurs?

Using path "/data/models/mlc/dist/models/Llama-3.2-3B-Instruct" for model "Llama-3.2-3B-Instruct"
Target configured: cuda -keys=cuda,gpu -arch=sm_87 -max_num_threads=1024 -max_shared_memory_per_block=49152 -max_threads_per_block=1024 -registers_per_block=65536 -thread_warp_size=32
Automatically using target for weight quantization: cuda -keys=cuda,gpu -arch=sm_87 -max_num_threads=1024 -max_shared_memory_per_block=49152 -max_threads_per_block=1024 -registers_per_block=65536 -thread_warp_size=32
Get old param:   0%|                                                                                       | 0/173 [00:00<?, ?tensors/sStart computing and quantizing weights... This may take a while.                                            | 0/287 [00:00<?, ?tensors/s]
Get old param:   1%|▍                                                                              | 1/173 [00:02<07:10,  2.50s/tensors]Traceback (most recent call last):
  File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/lib/python3.10/runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "/usr/local/lib/python3.10/dist-packages/mlc_llm/build.py", line 47, in <module>
    main()
  File "/usr/local/lib/python3.10/dist-packages/mlc_llm/build.py", line 43, in main
    core.build_model_from_args(parsed_args)
  File "/usr/local/lib/python3.10/dist-packages/mlc_llm/core.py", line 909, in build_model_from_args
    params = utils.convert_weights(mod_transform, param_manager, params, args)
  File "/usr/local/lib/python3.10/dist-packages/mlc_llm/utils.py", line 285, in convert_weights
    vm["transform_params"]()
  File "tvm/_ffi/_cython/./packed_func.pxi", line 332, in tvm._ffi._cy3.core.PackedFuncBase.__call__
  File "tvm/_ffi/_cython/./packed_func.pxi", line 263, in tvm._ffi._cy3.core.FuncCall
  File "tvm/_ffi/_cython/./packed_func.pxi", line 252, in tvm._ffi._cy3.core.FuncCall3
  File "tvm/_ffi/_cython/./base.pxi", line 182, in tvm._ffi._cy3.core.CHECK_CALL
  File "/usr/local/lib/python3.10/dist-packages/tvm/_ffi/base.py", line 481, in raise_last_ffi_error
    raise py_err
  File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
  File "/usr/local/lib/python3.10/dist-packages/mlc_llm/utils.py", line 48, in inner
    return func(*args, **kwargs)
  File "/usr/local/lib/python3.10/dist-packages/mlc_llm/relax_model/param_manager.py", line 622, in get_item
    for torch_binname in [
  File "/usr/local/lib/python3.10/dist-packages/mlc_llm/relax_model/param_manager.py", line 623, in <listcomp>
    self.torch_pname2binname[torch_pname] for torch_pname in torch_pnames
KeyError: 'lm_head.weight'
Traceback (most recent call last):
  File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/lib/python3.10/runpy.py", line 86, in _run_code
    exec(code, run_globals)
  File "/opt/NanoLLM/nano_llm/chat/__main__.py", line 32, in <module>
    model = NanoLLM.from_pretrained(
  File "/opt/NanoLLM/nano_llm/nano_llm.py", line 91, in from_pretrained
    model = MLCModel(model_path, **kwargs)
  File "/opt/NanoLLM/nano_llm/models/mlc.py", line 60, in __init__
    quant = MLCModel.quantize(self.model_path, self.config, method=quantization, max_context_len=max_context_len, **kwargs)
  File "/opt/NanoLLM/nano_llm/models/mlc.py", line 276, in quantize
    subprocess.run(cmd, executable='/bin/bash', shell=True, check=True)  
  File "/usr/lib/python3.10/subprocess.py", line 526, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command 'python3 -m mlc_llm.build --model /data/models/mlc/dist/models/Llama-3.2-3B-Instruct --quantization q4f16_ft --target cuda --use-cuda-graph --use-flash-attn-mqa --sep-embed --max-seq-len 131072 --artifact-path /data/models/mlc/dist/Llama-3.2-3B-Instruct/ctx131072 --use-safetensors ' returned non-zero exit status 1.

is there a list of supported llama models, trying get the latest versions