Instructions to use mozilla-ai/granite-34b-code-instruct-llamafile with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mozilla-ai/granite-34b-code-instruct-llamafile with Transformers:

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

pipe = pipeline("text-generation", model="mozilla-ai/granite-34b-code-instruct-llamafile")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("mozilla-ai/granite-34b-code-instruct-llamafile", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mozilla-ai/granite-34b-code-instruct-llamafile with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mozilla-ai/granite-34b-code-instruct-llamafile"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mozilla-ai/granite-34b-code-instruct-llamafile",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mozilla-ai/granite-34b-code-instruct-llamafile

SGLang

How to use mozilla-ai/granite-34b-code-instruct-llamafile 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 "mozilla-ai/granite-34b-code-instruct-llamafile" \
    --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": "mozilla-ai/granite-34b-code-instruct-llamafile",
		"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 "mozilla-ai/granite-34b-code-instruct-llamafile" \
        --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": "mozilla-ai/granite-34b-code-instruct-llamafile",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mozilla-ai/granite-34b-code-instruct-llamafile with Docker Model Runner:
```
docker model run hf.co/mozilla-ai/granite-34b-code-instruct-llamafile
```

jartine commited on May 26, 2024

Commit

5c09627

verified ·

1 Parent(s): 9f3578f

Update README.md

Browse files

Files changed (1) hide show

README.md +1 -3

README.md CHANGED Viewed

@@ -223,9 +223,7 @@ Assuming your system has at least 64GB of RAM, you can try running the
 following command which download, concatenate, and execute the model.
 ```
-( curl -L https://huggingface.co/jartine/granite-34b-code-instruct-llamafile/resolve/main/granite-34b-code-instruct.Q5_0.llamafile.cat0
-  curl -L https://huggingface.co/jartine/granite-34b-code-instruct-llamafile/resolve/main/granite-34b-code-instruct.Q5_0.llamafile.cat1
-) > granite-34b-code-instruct.Q5_0.llamafile
 chmod +x granite-34b-code-instruct.Q5_0.llamafile
 ./granite-34b-code-instruct.Q5_0.llamafile --help   # view manual
 ./granite-34b-code-instruct.Q5_0.llamafile          # launch web gui + oai api

 following command which download, concatenate, and execute the model.
 ```
+wget https://huggingface.co/jartine/granite-34b-code-instruct-llamafile/resolve/main/granite-34b-code-instruct.Q5_0.llamafile
 chmod +x granite-34b-code-instruct.Q5_0.llamafile
 ./granite-34b-code-instruct.Q5_0.llamafile --help   # view manual
 ./granite-34b-code-instruct.Q5_0.llamafile          # launch web gui + oai api