Instructions to use meta-llama/Llama-4-Scout-17B-16E with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use meta-llama/Llama-4-Scout-17B-16E with Transformers:

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

pipe = pipeline("image-text-to-text", model="meta-llama/Llama-4-Scout-17B-16E")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("meta-llama/Llama-4-Scout-17B-16E")
model = AutoModelForMultimodalLM.from_pretrained("meta-llama/Llama-4-Scout-17B-16E")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use meta-llama/Llama-4-Scout-17B-16E with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "meta-llama/Llama-4-Scout-17B-16E"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "meta-llama/Llama-4-Scout-17B-16E",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/meta-llama/Llama-4-Scout-17B-16E

SGLang

How to use meta-llama/Llama-4-Scout-17B-16E 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 "meta-llama/Llama-4-Scout-17B-16E" \
    --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": "meta-llama/Llama-4-Scout-17B-16E",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "meta-llama/Llama-4-Scout-17B-16E" \
        --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": "meta-llama/Llama-4-Scout-17B-16E",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use meta-llama/Llama-4-Scout-17B-16E with Docker Model Runner:
```
docker model run hf.co/meta-llama/Llama-4-Scout-17B-16E
```

Cannot generate with BS > 1

#25

by chenjiel - opened Apr 6, 2025

Discussion

chenjiel

Apr 6, 2025

Hi, Llama4 model developer, do you know how can we enable the HF transformers to load Llama4 scout and generate with BS > 1 inputs?

For example, I'm trying to test with the following code and will hit a shape error:

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "meta-llama/Llama-4-Scout-17B-16E-Instruct"

tokenizer = AutoTokenizer.from_pretrained(model_id)

input_texts = ["What's the age of the earth?", "What's the age of the sun?"]
input_ids = tokenizer.batch_encode_plus(
    input_texts,
    return_tensors="pt",
    padding=True,
    truncation=True)

model = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map="auto",
    torch_dtype=torch.bfloat16
)

outputs = model.generate(**input_ids.to(model.device), do_sample=False, max_new_tokens=200)
outputs = tokenizer.batch_decode(outputs)
print(outputs)

Error:

  File "/workspace/.local/lib/python3.12/site-packages/transformers/models/llama4/modeling_llama4.py", line 359, in forward
    attn_scales = attn_scales.view((*input_shape, 1, 1))
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: shape '[2, 7, 1, 1]' is invalid for input of size 7

Stack:

  File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/.local/lib/python3.12/site-packages/transformers/generation/utils.py", line 2460, in generate
    result = self._sample(
             ^^^^^^^^^^^^^
  File "/workspace/.local/lib/python3.12/site-packages/transformers/generation/utils.py", line 3426, in _sample
    outputs = self(**model_inputs, return_dict=True)
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1740, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1751, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/accelerate/hooks.py", line 176, in new_forward
    output = module._old_forward(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/.local/lib/python3.12/site-packages/transformers/models/llama4/modeling_llama4.py", line 1015, in forward
    outputs = self.model(
              ^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1740, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1751, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/.local/lib/python3.12/site-packages/transformers/models/llama4/modeling_llama4.py", line 700, in forward
    layer_outputs = decoder_layer(
                    ^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1740, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1751, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/accelerate/hooks.py", line 176, in new_forward
    output = module._old_forward(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/.local/lib/python3.12/site-packages/transformers/models/llama4/modeling_llama4.py", line 435, in forward
    attention_states, self_attn_weights = self.self_attn(
                                          ^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1740, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/nn/modules/module.py", line 1751, in _call_impl
    return forward_call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/accelerate/hooks.py", line 176, in new_forward
    output = module._old_forward(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspace/.local/lib/python3.12/site-packages/transformers/models/llama4/modeling_llama4.py", line 359, in forward
    attn_scales = attn_scales.view((*input_shape, 1, 1))
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: shape '[2, 7, 1, 1]' is invalid for input of size 7

ArthurZ

Apr 7, 2025

Fixed on main patching today!

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