Instructions to use Chunity/gemma-4-E4B-it-AWQ-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Chunity/gemma-4-E4B-it-AWQ-4bit with Transformers:

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

pipe = pipeline("image-text-to-text", model="Chunity/gemma-4-E4B-it-AWQ-4bit")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("Chunity/gemma-4-E4B-it-AWQ-4bit")
model = AutoModelForImageTextToText.from_pretrained("Chunity/gemma-4-E4B-it-AWQ-4bit")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Chunity/gemma-4-E4B-it-AWQ-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Chunity/gemma-4-E4B-it-AWQ-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Chunity/gemma-4-E4B-it-AWQ-4bit",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/Chunity/gemma-4-E4B-it-AWQ-4bit

SGLang

How to use Chunity/gemma-4-E4B-it-AWQ-4bit 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 "Chunity/gemma-4-E4B-it-AWQ-4bit" \
    --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": "Chunity/gemma-4-E4B-it-AWQ-4bit",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "Chunity/gemma-4-E4B-it-AWQ-4bit" \
        --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": "Chunity/gemma-4-E4B-it-AWQ-4bit",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use Chunity/gemma-4-E4B-it-AWQ-4bit with Docker Model Runner:
```
docker model run hf.co/Chunity/gemma-4-E4B-it-AWQ-4bit
```

gemma-4-E4B-it-AWQ-4bit / quantization_config.json

Chunity

Upload AutoRound AWQ 4-bit checkpoint

2a3cc32 verified about 1 month ago

raw

history blame contribute delete

4.8 kB

	{
	"bits": 4,
	"data_type": "int",
	"group_size": 128,
	"sym": true,
	"batch_size": 1,
	"iters": 500,
	"low_gpu_mem_usage": true,
	"nsamples": 32,
	"seqlen": 256,
	"to_quant_block_names": "model.language_model.layers",
	"autoround_version": "0.12.2",
	"provider": "auto-round",
	"quant_method": "awq",
	"zero_point": false,
	"version": "gemm",
	"modules_to_not_convert": [
	"model.embed_audio.embedding_projection",
	"model.audio_tower.subsample_conv_projection.input_proj_linear",
	"model.language_model.layers.31.self_attn.q_proj",
	"model.language_model.layers.23.self_attn.o_proj",
	"model.language_model.layers.34.self_attn.o_proj",
	"model.language_model.layers.27.self_attn.q_proj",
	"model.language_model.layers.36.self_attn",
	"model.language_model.layers.24.self_attn.o_proj",
	"model.language_model.layers.31.self_attn.o_proj",
	"model.language_model.layers.23.self_attn.q_proj",
	"model.language_model.layers.37.self_attn.q_proj",
	"model.language_model.layers.25.self_attn.o_proj",
	"model.language_model.layers.40.self_attn.o_proj",
	"model.language_model.layers.32.self_attn.q_proj",
	"model.language_model.layers.26.self_attn",
	"model.language_model.layers.5.self_attn.v_proj",
	"model.language_model.layers.25.self_attn",
	"model.language_model.layers.5.self_attn",
	"model.language_model.layers.23.self_attn.k_proj",
	"model.language_model.layers.30.self_attn",
	"model.language_model.layers.17.self_attn.v_proj",
	"model.language_model.layers.39.self_attn.o_proj",
	"lm_head",
	"model.language_model.layers.37.self_attn.o_proj",
	"model.language_model.layers.5.self_attn.k_proj",
	"model.language_model.layers.28.self_attn",
	"model.vision_tower.patch_embedder.input_proj",
	"model.language_model.layers.39.self_attn.q_proj",
	"model.language_model.layers.24.self_attn",
	"model.language_model.layers.35.self_attn.q_proj",
	"model.language_model.layers.29.self_attn.o_proj",
	"model.language_model.layers.31.self_attn",
	"model.language_model.layers.5.self_attn.q_proj",
	"model.language_model.layers.40.self_attn.q_proj",
	"model.language_model.layers.11.self_attn",
	"model.language_model.layers.29.self_attn",
	"model.language_model.layers.28.self_attn.q_proj",
	"model.language_model.layers.30.self_attn.o_proj",
	"model.language_model.layers.17.self_attn",
	"model.language_model.layers.17.self_attn.q_proj",
	"model.language_model.layers.11.self_attn.o_proj",
	"model.language_model.layers.28.self_attn.o_proj",
	"model.audio_tower.layers",
	"model.language_model.layers.23.self_attn.v_proj",
	"model.language_model.layers.25.self_attn.q_proj",
	"model.language_model.layers.26.self_attn.o_proj",
	"model.embed_vision",
	"model.language_model.per_layer_model_projection",
	"model.language_model.layers.27.self_attn.o_proj",
	"model.language_model.layers.37.self_attn",
	"model.language_model.layers.40.self_attn",
	"model.language_model.layers.36.self_attn.q_proj",
	"model.language_model.layers.32.self_attn.o_proj",
	"model.language_model.layers.27.self_attn",
	"model.language_model.layers.41.self_attn.q_proj",
	"model.language_model.layers.32.self_attn",
	"model.language_model.layers.30.self_attn.q_proj",
	"model.language_model.layers.34.self_attn.q_proj",
	"model.language_model.layers.23.self_attn",
	"model.language_model.layers.38.self_attn.o_proj",
	"model.language_model.layers.35.self_attn",
	"model.language_model.layers.33.self_attn.o_proj",
	"model.language_model.layers.35.self_attn.o_proj",
	"model.language_model.layers.29.self_attn.q_proj",
	"model.language_model.layers.17.self_attn.k_proj",
	"model.audio_tower",
	"model.language_model.layers.5.self_attn.o_proj",
	"model.audio_tower.output_proj",
	"model.language_model.layers.11.self_attn.q_proj",
	"model.language_model.layers.17.self_attn.o_proj",
	"model.language_model.layers.39.self_attn",
	"model.language_model.layers.11.self_attn.v_proj",
	"model.language_model.layers.41.self_attn.o_proj",
	"model.vision_tower",
	"model.language_model.layers.38.self_attn",
	"model.language_model.layers.38.self_attn.q_proj",
	"model.language_model.layers.33.self_attn.q_proj",
	"model.embed_vision.embedding_projection",
	"model.language_model.layers.41.self_attn",
	"model.embed_audio",
	"model.language_model.layers.26.self_attn.q_proj",
	"model.language_model.layers.36.self_attn.o_proj",
	"model.vision_tower.encoder.layers",
	"model.language_model.layers.24.self_attn.q_proj",
	"model.language_model.layers.34.self_attn",
	"model.language_model.layers.33.self_attn",
	"model.language_model.layers.11.self_attn.k_proj"
	]
	}