Instructions to use devpramod-intel/granite-4.1-8b-quantized.w8a8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use devpramod-intel/granite-4.1-8b-quantized.w8a8 with Transformers:

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

pipe = pipeline("text-generation", model="devpramod-intel/granite-4.1-8b-quantized.w8a8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("devpramod-intel/granite-4.1-8b-quantized.w8a8")
model = AutoModelForCausalLM.from_pretrained("devpramod-intel/granite-4.1-8b-quantized.w8a8", 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 devpramod-intel/granite-4.1-8b-quantized.w8a8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "devpramod-intel/granite-4.1-8b-quantized.w8a8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "devpramod-intel/granite-4.1-8b-quantized.w8a8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/devpramod-intel/granite-4.1-8b-quantized.w8a8

SGLang

How to use devpramod-intel/granite-4.1-8b-quantized.w8a8 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 "devpramod-intel/granite-4.1-8b-quantized.w8a8" \
    --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": "devpramod-intel/granite-4.1-8b-quantized.w8a8",
		"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 "devpramod-intel/granite-4.1-8b-quantized.w8a8" \
        --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": "devpramod-intel/granite-4.1-8b-quantized.w8a8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use devpramod-intel/granite-4.1-8b-quantized.w8a8 with Docker Model Runner:
```
docker model run hf.co/devpramod-intel/granite-4.1-8b-quantized.w8a8
```

devpramod-intel commited on 7 days ago

Commit

050ff66

verified ·

1 Parent(s): 303df57

Update model card

Browse files

Files changed (1) hide show

README.md +6 -0

README.md CHANGED Viewed

@@ -95,6 +95,12 @@ quant_stage:
       sequential_targets: [GraniteDecoderLayer]
 ```
 Calibration: `neuralmagic/LLM_compression_calibration`, `train` split,
 `shuffle(seed=42).select(512)`, the dataset's raw `text` field with
 `add_special_tokens=True`, `max_seq_length=8192`.

       sequential_targets: [GraniteDecoderLayer]
 ```
+`recipe.yaml` in this repo is what llm-compressor actually applied and is
+authoritative. It additionally shows `block_size: 128` and `actorder: static`,
+which are llm-compressor 0.9.0.4 defaults rather than choices — the older
+Granite cards predate `actorder` defaulting on, so this checkpoint is not
+bit-identical to what their recipe produced in 2025.
 Calibration: `neuralmagic/LLM_compression_calibration`, `train` split,
 `shuffle(seed=42).select(512)`, the dataset's raw `text` field with
 `add_special_tokens=True`, `max_seq_length=8192`.