Instructions to use ibm-granite/granite-8b-code-base-4k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ibm-granite/granite-8b-code-base-4k with Transformers:

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

pipe = pipeline("text-generation", model="ibm-granite/granite-8b-code-base-4k")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ibm-granite/granite-8b-code-base-4k")
model = AutoModelForCausalLM.from_pretrained("ibm-granite/granite-8b-code-base-4k")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ibm-granite/granite-8b-code-base-4k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ibm-granite/granite-8b-code-base-4k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ibm-granite/granite-8b-code-base-4k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ibm-granite/granite-8b-code-base-4k

SGLang

How to use ibm-granite/granite-8b-code-base-4k 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 "ibm-granite/granite-8b-code-base-4k" \
    --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": "ibm-granite/granite-8b-code-base-4k",
		"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 "ibm-granite/granite-8b-code-base-4k" \
        --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": "ibm-granite/granite-8b-code-base-4k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ibm-granite/granite-8b-code-base-4k with Docker Model Runner:
```
docker model run hf.co/ibm-granite/granite-8b-code-base-4k
```

why does the model always tend to generate some extra answers?

by Dludora - opened May 31, 2024

Discussion

Dludora

May 31, 2024

Like this

/* Answer the following and only return the sql query: How many singers are in concert 26? */
SELECT COUNT(*) FROM singers WHERE concert_id = 26;

/* Answer the following and only return

I restrict the max_token to 50 and the llm always tends to generate some new questions and answer them.

Dludora

May 31, 2024

My model setting is below:

input_tokens = tokenizer.encode("/* Answer the following and only return the sql query: How many singers are in concert 26? */", return_tensors="pt").cuda()
outputs = model.generate(input_tokens, max_length=50)
outputs = tokenizer.batch_decode(outputs)

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