Instructions to use alphaedge-ai/granite-4.0-h-350m-arb-16384 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use alphaedge-ai/granite-4.0-h-350m-arb-16384 with Transformers:

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

pipe = pipeline("text-generation", model="alphaedge-ai/granite-4.0-h-350m-arb-16384")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("alphaedge-ai/granite-4.0-h-350m-arb-16384")
model = AutoModelForCausalLM.from_pretrained("alphaedge-ai/granite-4.0-h-350m-arb-16384", device_map="auto")
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 Settings

vLLM

How to use alphaedge-ai/granite-4.0-h-350m-arb-16384 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "alphaedge-ai/granite-4.0-h-350m-arb-16384"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "alphaedge-ai/granite-4.0-h-350m-arb-16384",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/alphaedge-ai/granite-4.0-h-350m-arb-16384

SGLang

How to use alphaedge-ai/granite-4.0-h-350m-arb-16384 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 "alphaedge-ai/granite-4.0-h-350m-arb-16384" \
    --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": "alphaedge-ai/granite-4.0-h-350m-arb-16384",
		"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 "alphaedge-ai/granite-4.0-h-350m-arb-16384" \
        --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": "alphaedge-ai/granite-4.0-h-350m-arb-16384",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use alphaedge-ai/granite-4.0-h-350m-arb-16384 with Docker Model Runner:
```
docker model run hf.co/alphaedge-ai/granite-4.0-h-350m-arb-16384
```

granite-4.0-h-350m-arb-16384

This model is a 18.95% smaller version of ibm-granite/granite-4.0-h-350m optimized for Arabic language via vocabulary size reduction using the trimming method.
This trimmed model should perform similarly to the original model with only 16,384 tokens and a much smaller memory footprint. However, it may not perform well for other languages as tokens not commonly used in the selected languages were removed from the vocabulary.

Model Statistics

Metric	Original	Trimmed	Reduction
Vocabulary size	100,352 tokens	16,384 tokens	83.67%
Model size	340,332,224 params	275,844,800 params	18.95%

Mining Dataset Statistics

Number of texts used for mining: 200,000 texts
Dataset: lbourdois/fineweb-2-trimming

Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

device = "cuda"
model_path = "alphaedge-ai/granite-4.0-h-350m-arb-16384"

tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()

# change input text as desired
chat = [
    {"role": "user", "content": "Your prompt in Arabic."},
]
chat = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)

# tokenize the text
input_tokens = tokenizer(chat, return_tensors="pt").to(device)

# generate output tokens
output = model.generate(**input_tokens, max_new_tokens=100)

# decode output tokens into text
output = tokenizer.batch_decode(output)
print(output[0])

Citations

Granite 4.0

@misc{granite2025,
  author       = {IBM Research},
  title        = {Granite 4.0 Language Models},
  year         = {2025},
  howpublished = {https://github.com/ibm-granite/granite-4.0-language-models},
}

Trimming blog post

@misc{hf_blogpost_trimming,
      title={Introduction to Trimming}, 
      author={Loïck BOURDOIS and Tom AARSEN and Bram VANROY and Christopher AKIKI and Woojun JUNG and Manuel ROMERO and Prithiv SAKTHI},
      year={2026},
      url={https://huggingface.co/blog/lbourdois/introduction-to-trimming}, 
}