Instructions to use unsloth/granite-4.0-h-tiny-FP8-Dynamic with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use unsloth/granite-4.0-h-tiny-FP8-Dynamic with Transformers:

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

pipe = pipeline("text-generation", model="unsloth/granite-4.0-h-tiny-FP8-Dynamic")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("unsloth/granite-4.0-h-tiny-FP8-Dynamic")
model = AutoModelForCausalLM.from_pretrained("unsloth/granite-4.0-h-tiny-FP8-Dynamic")
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 unsloth/granite-4.0-h-tiny-FP8-Dynamic with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "unsloth/granite-4.0-h-tiny-FP8-Dynamic"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/granite-4.0-h-tiny-FP8-Dynamic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/unsloth/granite-4.0-h-tiny-FP8-Dynamic

SGLang

How to use unsloth/granite-4.0-h-tiny-FP8-Dynamic 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 "unsloth/granite-4.0-h-tiny-FP8-Dynamic" \
    --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": "unsloth/granite-4.0-h-tiny-FP8-Dynamic",
		"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 "unsloth/granite-4.0-h-tiny-FP8-Dynamic" \
        --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": "unsloth/granite-4.0-h-tiny-FP8-Dynamic",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio new

How to use unsloth/granite-4.0-h-tiny-FP8-Dynamic 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 unsloth/granite-4.0-h-tiny-FP8-Dynamic 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 unsloth/granite-4.0-h-tiny-FP8-Dynamic to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for unsloth/granite-4.0-h-tiny-FP8-Dynamic to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="unsloth/granite-4.0-h-tiny-FP8-Dynamic",
    max_seq_length=2048,
)

Docker Model Runner
How to use unsloth/granite-4.0-h-tiny-FP8-Dynamic with Docker Model Runner:
```
docker model run hf.co/unsloth/granite-4.0-h-tiny-FP8-Dynamic
```

danielhanchen commited on Nov 25, 2025

Commit

e079ee2

verified ·

1 Parent(s): 2cf226a

Add files using upload-large-folder tool

Browse files

Files changed (5) hide show

config.json +9 -5
generation_config.json +1 -1
model-00001-of-00002.safetensors +1 -1
model-00002-of-00002.safetensors +2 -2
model.safetensors.index.json +2 -3

config.json CHANGED Viewed

@@ -88,9 +88,11 @@
           "num_bits": 8,
           "observer": null,
           "observer_kwargs": {},
           "strategy": "token",
           "symmetric": true,
-          "type": "float"
         },
         "output_activations": null,
         "targets": [
@@ -105,9 +107,11 @@
           "num_bits": 8,
           "observer": "minmax",
           "observer_kwargs": {},
           "strategy": "channel",
           "symmetric": true,
-          "type": "float"
         }
       }
     },
@@ -161,7 +165,7 @@
     "quantization_status": "compressed",
     "sparsity_config": {},
     "transform_config": {},
-    "version": "0.12.2.a20251002"
   },
   "residual_multiplier": 0.22,
   "rms_norm_eps": 1e-05,
@@ -169,8 +173,8 @@
   "rope_theta": 10000,
   "router_aux_loss_coef": 0.0,
   "shared_intermediate_size": 1024,
-  "tie_word_embeddings": false,
-  "transformers_version": "4.57.0",
   "unsloth_fixed": true,
   "use_cache": true,
   "vocab_size": 100352

           "num_bits": 8,
           "observer": null,
           "observer_kwargs": {},
+          "scale_dtype": null,
           "strategy": "token",
           "symmetric": true,
+          "type": "float",
+          "zp_dtype": null
         },
         "output_activations": null,
         "targets": [
           "num_bits": 8,
           "observer": "minmax",
           "observer_kwargs": {},
+          "scale_dtype": null,
           "strategy": "channel",
           "symmetric": true,
+          "type": "float",
+          "zp_dtype": null
         }
       }
     },
     "quantization_status": "compressed",
     "sparsity_config": {},
     "transform_config": {},
+    "version": "0.12.3.a20251114"
   },
   "residual_multiplier": 0.22,
   "rms_norm_eps": 1e-05,
   "rope_theta": 10000,
   "router_aux_loss_coef": 0.0,
   "shared_intermediate_size": 1024,
+  "tie_word_embeddings": true,
+  "transformers_version": "4.57.1",
   "unsloth_fixed": true,
   "use_cache": true,
   "vocab_size": 100352

generation_config.json CHANGED Viewed

@@ -4,5 +4,5 @@
   "eos_token_id": 100257,
   "max_length": 131072,
   "pad_token_id": 100256,
-  "transformers_version": "4.57.0"
 }

   "eos_token_id": 100257,
   "max_length": 131072,
   "pad_token_id": 100256,
+  "transformers_version": "4.57.1"
 }

model-00001-of-00002.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9d6e55462bbb95b5b14f52de8facc1d7b1fe5d16f047c9eeb7c13ae64835e60
 size 4999103976

 version https://git-lfs.github.com/spec/v1
+oid sha256:e80671bd29d8523c598fa21fb569c79ae06e3102f701f27f326cccd5a7f35917
 size 4999103976

model-00002-of-00002.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37778dba5a60ce8eca29fa432fe933b012c4d508388b9e4aa29dceba4537f192
-size 2421227416

 version https://git-lfs.github.com/spec/v1
+oid sha256:963e7dcb29b2a8249147a492b1650b108695a71ee01f1d6615c18767ad183fd6
+size 2112945144

model.safetensors.index.json CHANGED Viewed

@@ -1,10 +1,9 @@
 {
   "metadata": {
-    "total_parameters": 7100178688,
-    "total_size": 7420232192
   },
   "weight_map": {
-    "lm_head.weight": "model-00002-of-00002.safetensors",
     "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
     "model.layers.0.block_sparse_moe.input_linear.weight": "model-00001-of-00002.safetensors",
     "model.layers.0.block_sparse_moe.input_linear.weight_scale": "model-00001-of-00002.safetensors",

 {
   "metadata": {
+    "total_parameters": 6946038016,
+    "total_size": 7111950848
   },
   "weight_map": {
     "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
     "model.layers.0.block_sparse_moe.input_linear.weight": "model-00001-of-00002.safetensors",
     "model.layers.0.block_sparse_moe.input_linear.weight_scale": "model-00001-of-00002.safetensors",