Instructions to use mlx-community/granite-3b-code-base-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlx-community/granite-3b-code-base-8bit with Transformers:

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

pipe = pipeline("text-generation", model="mlx-community/granite-3b-code-base-8bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mlx-community/granite-3b-code-base-8bit")
model = AutoModelForCausalLM.from_pretrained("mlx-community/granite-3b-code-base-8bit")

MLX

How to use mlx-community/granite-3b-code-base-8bit with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm
# if on a CUDA device, also pip install mlx[cuda]

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/granite-3b-code-base-8bit")

prompt = "Once upon a time in"
text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps
LM Studio

vLLM

How to use mlx-community/granite-3b-code-base-8bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlx-community/granite-3b-code-base-8bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlx-community/granite-3b-code-base-8bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mlx-community/granite-3b-code-base-8bit

SGLang

How to use mlx-community/granite-3b-code-base-8bit 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 "mlx-community/granite-3b-code-base-8bit" \
    --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": "mlx-community/granite-3b-code-base-8bit",
		"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 "mlx-community/granite-3b-code-base-8bit" \
        --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": "mlx-community/granite-3b-code-base-8bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

MLX LM

How to use mlx-community/granite-3b-code-base-8bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Generate some text
mlx_lm.generate --model "mlx-community/granite-3b-code-base-8bit" --prompt "Once upon a time"

Docker Model Runner
How to use mlx-community/granite-3b-code-base-8bit with Docker Model Runner:
```
docker model run hf.co/mlx-community/granite-3b-code-base-8bit
```

Upload folder using huggingface_hub

by prince-canuma - opened May 6, 2024

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+148510

-0

prince-canuma

MLX Community org May 6, 2024

•

edited May 6, 2024

Upload folder using huggingface_hub

Multi commit ID: 8d7f078615d278e9873075e97d3f8183f2ea6e83a50152ac82c2e00d6a0eee94

Scheduled commits:

Upload 1 file(s) totalling 3.7G (e7115d76d1b88df9826346a12c594113dd1b888560bc12219722fe93ea21b09a)
Upload 8 file(s) totalling 3.4M (d99937627d5c7aa23c19cf14b62669664a0b77ab061afe03be9cc3fdf74f3efa)

This is a PR opened using the huggingface_hub library in the context of a multi-commit. PR can be commented as a usual PR. However, please be aware that manually updating the PR description, changing the PR status, or pushing new commits, is not recommended as it might corrupt the commit process. Learn more about multi-commits in this guide.

e7115d76d1b88df9826346a12c594113dd1b888560bc12219722fe93ea21b09a4b190a41

d99937627d5c7aa23c19cf14b62669664a0b77ab061afe03be9cc3fdf74f3efa386cbba3

prince-canuma changed pull request title from [WIP] Upload folder using huggingface_hub (multi-commit 8d7f078615d278e9873075e97d3f8183f2ea6e83a50152ac82c2e00d6a0eee94) to Upload folder using huggingface_hub May 6, 2024

prince-canuma

MLX Community org May 6, 2024

Multi-commit is now completed! You can ping the repo owner to review the changes. This PR can now be commented or modified without risking to corrupt it.

This is a comment posted using the huggingface_hub library in the context of a multi-commit. Learn more about multi-commits in this guide.

prince-canuma changed pull request status to open May 6, 2024

prince-canuma

MLX Community org May 6, 2024

create_pr=False has been passed so PR is automatically merged.

This is a comment posted using the huggingface_hub library in the context of a multi-commit. Learn more about multi-commits in this guide.

prince-canuma changed pull request status to merged May 6, 2024

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