Instructions to use itzune/gemma-4-e4b-horkonpon with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use itzune/gemma-4-e4b-horkonpon with Transformers:

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

pipe = pipeline("text-generation", model="itzune/gemma-4-e4b-horkonpon")
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, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("itzune/gemma-4-e4b-horkonpon")
model = AutoModelForMultimodalLM.from_pretrained("itzune/gemma-4-e4b-horkonpon", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use itzune/gemma-4-e4b-horkonpon with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "itzune/gemma-4-e4b-horkonpon"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "itzune/gemma-4-e4b-horkonpon",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/itzune/gemma-4-e4b-horkonpon

SGLang

How to use itzune/gemma-4-e4b-horkonpon 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 "itzune/gemma-4-e4b-horkonpon" \
    --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": "itzune/gemma-4-e4b-horkonpon",
		"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 "itzune/gemma-4-e4b-horkonpon" \
        --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": "itzune/gemma-4-e4b-horkonpon",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use itzune/gemma-4-e4b-horkonpon with Docker Model Runner:
```
docker model run hf.co/itzune/gemma-4-e4b-horkonpon
```

xezpeleta commited on 2 days ago

Commit

ee348a6

verified ·

1 Parent(s): b89dc6b

Add base model ablation eval report

Browse files

Files changed (1) hide show

eval_report_base.json +79 -0

eval_report_base.json ADDED Viewed

	@@ -0,0 +1,79 @@

+{
+  "model": "unsloth/gemma-4-e4b-it",
+  "load_4bit": true,
+  "eval_file": "data/sft/eval.jsonl",
+  "n_errorful": 1088,
+  "n_clean": 1088,
+  "json_parse_rate": 92.83088235294117,
+  "exact_match": 76,
+  "exact_match_pct": 6.985294117647059,
+  "tp": 193,
+  "fp": 10701,
+  "fn": 976,
+  "precision": 1.7716174040756378,
+  "recall": 16.5098374679213,
+  "f05": 2.1566655492233764,
+  "unchanged_error": 195,
+  "changed_wrong": 817,
+  "per_category_recall": {
+    "morphology": {
+      "tp": 236,
+      "total": 292,
+      "recall": 80.82191780821918
+    },
+    "spelling": {
+      "tp": 138,
+      "total": 170,
+      "recall": 81.17647058823529
+    },
+    "punctuation": {
+      "tp": 215,
+      "total": 258,
+      "recall": 83.33333333333334
+    },
+    "capitalization": {
+      "tp": 33,
+      "total": 33,
+      "recall": 100.0
+    },
+    "word_level": {
+      "tp": 33,
+      "total": 36,
+      "recall": 91.66666666666666
+    },
+    "zalantza": {
+      "tp": 63,
+      "total": 89,
+      "recall": 70.78651685393258
+    },
+    "proper_noun": {
+      "tp": 73,
+      "total": 83,
+      "recall": 87.95180722891565
+    },
+    "calque": {
+      "tp": 56,
+      "total": 76,
+      "recall": 73.68421052631578
+    },
+    "terminology": {
+      "tp": 46,
+      "total": 51,
+      "recall": 90.19607843137256
+    }
+  },
+  "per_nature_recall": {
+    "error": {
+      "tp": 847,
+      "total": 1037,
+      "recall": 81.67791706846673
+    },
+    "suggestion": {
+      "tp": 46,
+      "total": 51,
+      "recall": 90.19607843137256
+    }
+  },
+  "clean_fp": 1058,
+  "clean_fp_pct": 97.24264705882352
+}