Instructions to use ibm-granite/granite-20b-code-instruct-8k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ibm-granite/granite-20b-code-instruct-8k with Transformers:

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

pipe = pipeline("text-generation", model="ibm-granite/granite-20b-code-instruct-8k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ibm-granite/granite-20b-code-instruct-8k")
model = AutoModelForCausalLM.from_pretrained("ibm-granite/granite-20b-code-instruct-8k")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ibm-granite/granite-20b-code-instruct-8k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ibm-granite/granite-20b-code-instruct-8k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ibm-granite/granite-20b-code-instruct-8k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ibm-granite/granite-20b-code-instruct-8k

SGLang

How to use ibm-granite/granite-20b-code-instruct-8k 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-20b-code-instruct-8k" \
    --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": "ibm-granite/granite-20b-code-instruct-8k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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-20b-code-instruct-8k" \
        --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": "ibm-granite/granite-20b-code-instruct-8k",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ibm-granite/granite-20b-code-instruct-8k with Docker Model Runner:
```
docker model run hf.co/ibm-granite/granite-20b-code-instruct-8k
```

Is it fine to do granite-20b model's inference with bfloat16 dtype?

by lkm1 - opened Jun 2, 2024

Discussion

lkm1

Jun 2, 2024

I have tried this model's inference with bfloat16 type and have not faced any problems. Is it fine to use bfloat16 dtype for this model's inference, without much effect on score?

mayank-mishra

Jun 4, 2024

yeah, it should be ok @lkm1

mayank-mishra

Jun 4, 2024

we have run scores in bf16

lkm1

Jun 4, 2024

ok, thanks @mayank-mishra !

lkm1 changed discussion status to closed Jun 4, 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