Instructions to use nvidia/Llama-3_3-Nemotron-Super-49B-v1_5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nvidia/Llama-3_3-Nemotron-Super-49B-v1_5 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="nvidia/Llama-3_3-Nemotron-Super-49B-v1_5", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("nvidia/Llama-3_3-Nemotron-Super-49B-v1_5", trust_remote_code=True, dtype="auto")

Inference
Local Apps Settings

vLLM

How to use nvidia/Llama-3_3-Nemotron-Super-49B-v1_5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nvidia/Llama-3_3-Nemotron-Super-49B-v1_5

SGLang

How to use nvidia/Llama-3_3-Nemotron-Super-49B-v1_5 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3_3-Nemotron-Super-49B-v1_5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nvidia/Llama-3_3-Nemotron-Super-49B-v1_5 with Docker Model Runner:
```
docker model run hf.co/nvidia/Llama-3_3-Nemotron-Super-49B-v1_5
```

cannot import name 'NEED_SETUP_CACHE_CLASSES_MAPPING'

#11

by uygarkurt - opened Nov 23, 2025

Discussion

uygarkurt

Nov 23, 2025

Hi. Currently, I can load this model using HuggingFace from_pretrained with the transformers==4.44.2.

However, I need to use it with transformers>=4.53.0. When I do that, I get the following:

model = AutoModelForCausalLM.from_pretrained(
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File ".venv/lib/python3.12/site-packages/transformers/models/auto/auto_factory.py", line 586, in from_pretrained
    model_class = get_class_from_dynamic_module(
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "home.venv/lib/python3.12/site-packages/transformers/dynamic_module_utils.py", line 581, in get_class_from_dynamic_module
    return get_class_in_module(class_name, final_module, force_reload=force_download)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "home.venv/lib/python3.12/site-packages/transformers/dynamic_module_utils.py", line 276, in get_class_in_module
    module_spec.loader.exec_module(module)
  File "<frozen importlib._bootstrap_external>", line 995, in exec_module
  File "<frozen importlib._bootstrap>", line 488, in _call_with_frames_removed
  File "/home/.cache/huggingface/modules/transformers_modules/Llama-3_3-Nemotron-Super-49B-v1_5/modeling_decilm.py", line 30, in <module>
    from transformers.generation.utils import NEED_SETUP_CACHE_CLASSES_MAPPING, GenerationMixin, GenerateOutput
ImportError: cannot import name 'NEED_SETUP_CACHE_CLASSES_MAPPING' from 'transformers.generation.utils' (home.venv/lib/python3.12/site-packages/transformers/generation/utils.py)

Is there any solution to it? Thank you.

uygarkurt changed discussion title from transformers>=4.53.0 Support to cannot import name 'NEED_SETUP_CACHE_CLASSES_MAPPING' Nov 23, 2025

NePe

Nov 24, 2025

You can just delete the NEED_SETUP_CACHE_CLASSES_MAPPING references from the code. But then there will be no caching so decoding will be slow. I did some modifications and the caching seems to be working with the latest transformers version. My code is a bit messy right now, maybe i will do a pr later this week.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment