Instructions to use ibm-granite/granite-4.1-8b-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ibm-granite/granite-4.1-8b-base with Transformers:

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

pipe = pipeline("text-generation", model="ibm-granite/granite-4.1-8b-base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("ibm-granite/granite-4.1-8b-base")
model = AutoModelForMultimodalLM.from_pretrained("ibm-granite/granite-4.1-8b-base")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ibm-granite/granite-4.1-8b-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ibm-granite/granite-4.1-8b-base"
# 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-4.1-8b-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ibm-granite/granite-4.1-8b-base

SGLang

How to use ibm-granite/granite-4.1-8b-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 "ibm-granite/granite-4.1-8b-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": "ibm-granite/granite-4.1-8b-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 "ibm-granite/granite-4.1-8b-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": "ibm-granite/granite-4.1-8b-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ibm-granite/granite-4.1-8b-base with Docker Model Runner:
```
docker model run hf.co/ibm-granite/granite-4.1-8b-base
```

Base model

by Ethermich - opened Apr 30

Discussion

Ethermich

Apr 30

Dear team,

Could you please share more information regarding the base model training?
The intended use case we are evaluating for is fine tuning for writing tasks (fiction, non fiction).
Would you consider the Granite 4.1 a good base for such use case?
Would a o fine-tuning recipe developed on the 8b be applied relatively smoothly to the 30b?

Thanks in advance.

Thanks,
Michael

gabegoodhart

IBM Granite org Apr 30

Hi Michael, you can find all the training details in the technical blog: https://huggingface.co/blog/ibm-granite/granite-4-1

Ethermich

May 2

So far going well. I wonder though why did you pivot back to dense models?
We haven’t had issues fine tuning moes for our use case. In terms of quality it gave as good results as dense variants while delivering much higher inference speeds…

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