Instructions to use nvidia/Llama-3.1-Minitron-4B-Width-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nvidia/Llama-3.1-Minitron-4B-Width-Base with Transformers:

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

pipe = pipeline("text-generation", model="nvidia/Llama-3.1-Minitron-4B-Width-Base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nvidia/Llama-3.1-Minitron-4B-Width-Base")
model = AutoModelForCausalLM.from_pretrained("nvidia/Llama-3.1-Minitron-4B-Width-Base", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nvidia/Llama-3.1-Minitron-4B-Width-Base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nvidia/Llama-3.1-Minitron-4B-Width-Base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3.1-Minitron-4B-Width-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nvidia/Llama-3.1-Minitron-4B-Width-Base

SGLang

How to use nvidia/Llama-3.1-Minitron-4B-Width-Base 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-3.1-Minitron-4B-Width-Base" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3.1-Minitron-4B-Width-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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-3.1-Minitron-4B-Width-Base" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Llama-3.1-Minitron-4B-Width-Base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nvidia/Llama-3.1-Minitron-4B-Width-Base with Docker Model Runner:
```
docker model run hf.co/nvidia/Llama-3.1-Minitron-4B-Width-Base
```

Weight Error in Notebook

by atharvanighot - opened Aug 18, 2024

Discussion

atharvanighot

Aug 18, 2024

Getting this error:

ValueError: Trying to set a tensor of shape torch.Size([1024, 3072]) in "weight" (which has shape torch.Size([768, 3072])), this look incorrect.

while running default script:

import torch
from transformers import AutoTokenizer, LlamaForCausalLM

# Load the tokenizer and model
model_path = "nvidia/Llama3.1-Minitron-4B-Width-Base"
tokenizer = AutoTokenizer.from_pretrained(model_path)

device = 'cuda'
dtype = torch.bfloat16
model = LlamaForCausalLM.from_pretrained(model_path, torch_dtype=dtype, device_map=device)

# Prepare the input text
prompt = 'Complete the paragraph: our solar system is'
inputs = tokenizer.encode(prompt, return_tensors='pt').to(model.device)

# Generate the output
outputs = model.generate(inputs, max_length=20)

# Decode and print the output
output_text = tokenizer.decode(outputs[0])
print(output_text)

KevinStorm86

Aug 19, 2024

•

edited Aug 19, 2024

probably issue on GroudQueryAttention

KevinStorm86

Aug 19, 2024

@atharvanighot

The pull requests to support this model in Hugging Face Transformers are currently under review.
Follow the installation instructions below:

Fetch PR 32502

$ git clone -b suhara/llama-kv-channels --single-branch https://github.com/suhara/transformers.git && cd transformers

Fetch changes from PR 32495

$ git fetch https://github.com/suiyoubi/transformers.git aot/head_dim_rope && git cherry-pick FETCH_HEAD --strategy-option theirs

Install transformers

$ pip install -e

dchaplinsky

Aug 20, 2024

Will subscribe to that

srvm

NVIDIA org Aug 22, 2024

@atharvanighot are you still getting this error? The installation instructions have been updated - you no longer need to fetch these PRs manually:

pip install git+https://github.com/huggingface/transformers

atharvanighot

Aug 22, 2024

@srvm

I tried it once again without manually fetching the PRs. I made sure to upgrade transformers, but I'm still getting this error:

ValueError: Trying to set a tensor of shape torch.Size([1024, 3072]) in "weight" (which has shape torch.Size([768, 3072])), this look incorrect.

I'll try it again later by fetching PRs manually.

Tiz01

Aug 23, 2024

same error

atharvanighot changed discussion status to closed Aug 23, 2024

atharvanighot changed discussion status to open Aug 23, 2024

Zephyros

Aug 27, 2024

@atharvanighot , @Tiz01
Though Depth prunned model works fine. You could try using it instead

mobeidat

Aug 31, 2024

Still same issue today with latest transformer package.

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