Instructions to use jstarks/gemma-4-E4B-it-weight-space-programming with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jstarks/gemma-4-E4B-it-weight-space-programming with Transformers:

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

pipe = pipeline("image-text-to-text", model="jstarks/gemma-4-E4B-it-weight-space-programming")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("jstarks/gemma-4-E4B-it-weight-space-programming")
model = AutoModelForMultimodalLM.from_pretrained("jstarks/gemma-4-E4B-it-weight-space-programming", device_map="auto")
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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jstarks/gemma-4-E4B-it-weight-space-programming with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jstarks/gemma-4-E4B-it-weight-space-programming"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jstarks/gemma-4-E4B-it-weight-space-programming",
		"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/jstarks/gemma-4-E4B-it-weight-space-programming

SGLang

How to use jstarks/gemma-4-E4B-it-weight-space-programming 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 "jstarks/gemma-4-E4B-it-weight-space-programming" \
    --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": "jstarks/gemma-4-E4B-it-weight-space-programming",
		"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 "jstarks/gemma-4-E4B-it-weight-space-programming" \
        --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": "jstarks/gemma-4-E4B-it-weight-space-programming",
		"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 jstarks/gemma-4-E4B-it-weight-space-programming with Docker Model Runner:
```
docker model run hf.co/jstarks/gemma-4-E4B-it-weight-space-programming
```

gemma-4-E4B-it-weight-space-programming

File size: 1,176 Bytes

576ab11

"""
Apply the 64-column Weight-Space Patch (`gemma4_e4b_weight_space_patch.safetensors`)
to any loaded `Gemma4ForCausalLM` or `Gemma4ForConditionalGeneration` instance in <1 second!
"""

import torch
from safetensors.torch import load_file


@torch.no_grad()
def apply_gemma4_weight_space_patch(model, patch_safetensors_path: str = "gemma4_e4b_weight_space_patch.safetensors"):
    tensors = load_file(patch_safetensors_path, device="cpu")
    # Support both Gemma4ForCausalLM (model.model.layers) and Gemma4ForConditionalGeneration (model.model.language_model.layers)
    layers = getattr(model.model, "layers", None)
    if layers is None:
        layers = model.model.language_model.layers

    total_params = 0
    for li in [38, 39, 40, 41]:
        cols = tensors[f"layers.{li}.cols"].long()
        patched_cols = tensors[f"layers.{li}.patched_columns"]
        W = layers[li].mlp.down_proj.weight.data
        W[:, cols.to(W.device)] = patched_cols.to(device=W.device, dtype=W.dtype)
        total_params += patched_cols.numel()

    print(f"Applied Gemma-4-E4B-it weight-space patch ({total_params:,} parameters across 64 columns in Layers 38-41).")
    return model