Instructions to use CountingSheep/vev-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CountingSheep/vev-4b with Transformers:

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

pipe = pipeline("image-text-to-text", model="CountingSheep/vev-4b")
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("CountingSheep/vev-4b")
model = AutoModelForMultimodalLM.from_pretrained("CountingSheep/vev-4b", 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 CountingSheep/vev-4b with vLLM:

Install from pip and serve model

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

SGLang

How to use CountingSheep/vev-4b 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 "CountingSheep/vev-4b" \
    --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": "CountingSheep/vev-4b",
		"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 "CountingSheep/vev-4b" \
        --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": "CountingSheep/vev-4b",
		"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 CountingSheep/vev-4b with Docker Model Runner:
```
docker model run hf.co/CountingSheep/vev-4b
```

CountingSheep commited on 29 days ago

Commit

8b98b2a

verified ·

1 Parent(s): 0dcd077

Card: replace the trust section with Limitations

Browse files

Files changed (1) hide show

README.md +7 -6

README.md CHANGED Viewed

@@ -101,14 +101,15 @@ Accuracy on human-labelled data, by the kind of question:
 | General questions about a photo | MMBench-EN / POPE / MMStar | 0.881 / 0.889 / 0.627 |
 | Bugs in game screenshots | glitch / object clipping (VideoGameQA-Bench) | 0.613 / 0.671 |
-The text rows are public benchmarks; the image rows are judgment sets converted from labelled public data. Check it
-on a labelled sample of your own questions before relying on it. Significance tests and the comparison with the
-base model are in the [GitHub README](https://github.com/Xiaooolong/vev#detailed-results).
-## How far to trust it
-- Jev is more accurate on nimble (by 22 points), JevBench (10) and kev transfer-v4 (8); they tie on judgekit. The
-  one significant regression from fine-tuning is POPE (0.5 points lower than Qwen3.5-4B).
 - The probabilities rank answers well but are not exact frequencies: expected calibration error is 0.02–0.11
   depending on the task. Choose any threshold on your own labelled data.
 - Reversing the order of the options changes the top answer on 17% of JevBench and kev transfer-v4 questions (Jev:

 | General questions about a photo | MMBench-EN / POPE / MMStar | 0.881 / 0.889 / 0.627 |
 | Bugs in game screenshots | glitch / object clipping (VideoGameQA-Bench) | 0.613 / 0.671 |
+The text rows are public benchmarks; the image rows are judgment sets converted from labelled public data. Jev is
+more accurate on nimble (by 22 points), JevBench (10) and kev transfer-v4 (8); they tie on judgekit. Jev's API
+takes text only, so the image rows have no Jev comparison. The one significant regression from fine-tuning is POPE
+(0.5 points lower than Qwen3.5-4B). Check Vev on a labelled sample of your own questions before relying on it.
+Significance tests and the comparison with the base model are in the [GitHub
+README](https://github.com/Xiaooolong/vev#detailed-results).
+## Limitations
 - The probabilities rank answers well but are not exact frequencies: expected calibration error is 0.02–0.11
   depending on the task. Choose any threshold on your own labelled data.
 - Reversing the order of the options changes the top answer on 17% of JevBench and kev transfer-v4 questions (Jev: