Instructions to use DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X with Transformers:

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

pipe = pipeline("text-generation", model="DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X")
model = AutoModelForCausalLM.from_pretrained("DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X")
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 DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X

SGLang

How to use DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X 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 "DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X" \
    --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": "DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X",
		"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 "DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X" \
        --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": "DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X",
    max_seq_length=2048,
)

Docker Model Runner
How to use DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X with Docker Model Runner:
```
docker model run hf.co/DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Granite-4.1-30B-Claude-4.6-Opus-Thinking-X

Fine tune of Granite 4.1 with Unsloth on local hardware to convert model from "instruct" to full "reasoning/thinking".

Other versions below.

Context: 128k

Suggest: Temp 1, Topp: .95, Minp .05, rep pen 1 (off) OR for creative 1.05 to 1.1, min context of 8k.

Enjoy ;

TECH NOTE:

The "lm_head" was split from the "embed" to improve training and quant performance prior to training.

For GGUF quants, this will allow you to set the "output tensor" at bf16 and get stronger performance overall.

BENCHMARKS by Nightmedia:

         arc-c arc/e boolq hswag obkqa piqa  wino

Granite-4.1-30B-Claude-4.6-Opus-Thinking-Charles-Xavier
mxfp8    0.573,0.761,0.876,...

Granite-4.1-30B-Claude-4.6-Opus-Thinking-Xavier
mxfp8    0.563,0.739,0.879,0.722,0.430,0.779,0.723

THIS MODEL:
Granite-4.1-30B-Claude-4.6-Opus-Thinking-X
qx64-hi  0.526,0.696,0.894,...

- BASE UNTUNED MODEL -

granite-4.1-30b
mxfp8    0.456,0.572,0.897,0.621,0.444,0.757,0.616

granite-4.1-30b
qx64-hi  0.462,0.582,0.896,0.642,0.448,0.769,0.600

Downloads last month: 39

Safetensors

Model size

29B params

Tensor type

BF16

Model tree for DavidAU/Granite-4.1-30B-Claude-4.6-Opus-Thinking-X

Base model

ibm-granite/granite-4.1-30b

Finetuned

(8)

this model

Merges

8 models

Quantizations

2 models