Instructions to use voidful/Qwen3.5-9B-earica with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use voidful/Qwen3.5-9B-earica with Transformers:

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

pipe = pipeline("image-text-to-text", model="voidful/Qwen3.5-9B-earica")
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("voidful/Qwen3.5-9B-earica")
model = AutoModelForMultimodalLM.from_pretrained("voidful/Qwen3.5-9B-earica", 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 voidful/Qwen3.5-9B-earica with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "voidful/Qwen3.5-9B-earica"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "voidful/Qwen3.5-9B-earica",
		"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/voidful/Qwen3.5-9B-earica

SGLang

How to use voidful/Qwen3.5-9B-earica 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 "voidful/Qwen3.5-9B-earica" \
    --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": "voidful/Qwen3.5-9B-earica",
		"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 "voidful/Qwen3.5-9B-earica" \
        --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": "voidful/Qwen3.5-9B-earica",
		"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 voidful/Qwen3.5-9B-earica with Docker Model Runner:
```
docker model run hf.co/voidful/Qwen3.5-9B-earica
```

Qwen3.5-9B-earica / debug.log

voidful

End of training

7c913da verified 5 months ago

Raw

History Blame Contribute Delete

3.38 kB

	Fetching 22 files: 0%\| \| 0/22 [00:00<?, ?it/s] Fetching 22 files: 100%\|██████████\| 22/22 [00:00<00:00, 59302.50it/s]
	Loading weights: 0%\| \| 0/760 [00:00<?, ?it/s] Loading weights: 0%\| \| 1/760 [00:01<19:03, 1.51s/it] Loading weights: 2%\|▎ \| 19/760 [00:01<00:45, 16.20it/s] Loading weights: 4%\|▍ \| 30/760 [00:01<00:27, 26.34it/s] Loading weights: 6%\|▌ \| 43/760 [00:01<00:17, 40.21it/s] Loading weights: 7%\|▋ \| 55/760 [00:01<00:13, 52.83it/s] Loading weights: 9%\|▉ \| 68/760 [00:02<00:10, 66.68it/s] Loading weights: 11%\|█ \| 82/760 [00:02<00:08, 81.99it/s] Loading weights: 13%\|█▎ \| 96/760 [00:02<00:06, 94.98it/s] Loading weights: 15%\|█▍ \| 112/760 [00:02<00:05, 109.47it/s] Loading weights: 17%\|█▋ \| 129/760 [00:02<00:05, 121.97it/s] Loading weights: 19%\|█▉ \| 144/760 [00:02<00:04, 127.16it/s] Loading weights: 21%\|██ \| 159/760 [00:02<00:06, 98.70it/s] Loading weights: 23%\|██▎ \| 174/760 [00:02<00:05, 104.88it/s] Loading weights: 24%\|██▍ \| 186/760 [00:03<00:05, 101.19it/s] Loading weights: 26%\|██▌ \| 199/760 [00:03<00:05, 102.44it/s] Loading weights: 28%\|██▊ \| 210/760 [00:03<00:06, 81.72it/s] Loading weights: 30%\|██▉ \| 225/760 [00:03<00:05, 93.85it/s] Loading weights: 31%\|███ \| 236/760 [00:03<00:05, 94.65it/s] Loading weights: 33%\|███▎ \| 250/760 [00:03<00:04, 103.57it/s] Loading weights: 34%\|███▍ \| 262/760 [00:03<00:05, 90.67it/s] Loading weights: 37%\|███▋ \| 280/760 [00:03<00:04, 106.52it/s] Loading weights: 38%\|███▊ \| 292/760 [00:04<00:04, 101.22it/s] Loading weights: 40%\|████ \| 306/760 [00:04<00:04, 107.57it/s] Loading weights: 42%\|████▏ \| 318/760 [00:04<00:05, 78.15it/s] Loading weights: 44%\|████▎ \| 331/760 [00:04<00:05, 81.83it/s] Loading weights: 46%\|████▌ \| 346/760 [00:04<00:04, 89.24it/s] Loading weights: 47%\|████▋ \| 360/760 [00:04<00:04, 89.88it/s] Loading weights: 49%\|████▊ \| 370/760 [00:05<00:04, 82.66it/s] Loading weights: 50%\|█████ \| 380/760 [00:05<00:04, 77.44it/s] Loading weights: 51%\|█████▏ \| 390/760 [00:05<00:04, 80.60it/s] Loading weights: 52%\|█████▎ \| 399/760 [00:05<00:04, 75.67it/s] Loading weights: 54%\|█████▍ \| 413/760 [00:05<00:03, 89.50it/s] Loading weights: 56%\|█████▌ \| 423/760 [00:05<00:04, 73.86it/s] Loading weights: 61%\|██████▏ \| 466/760 [00:05<00:01, 150.47it/s] Loading weights: 71%\|███████ \| 537/760 [00:06<00:00, 275.94it/s] Loading weights: 80%\|███████▉ \| 607/760 [00:06<00:00, 378.05it/s] Loading weights: 86%\|████████▌ \| 650/760 [00:06<00:00, 390.34it/s] Loading weights: 94%\|█████████▎\| 711/760 [00:06<00:00, 442.87it/s] Loading weights: 100%\|█████████▉\| 759/760 [00:06<00:00, 413.35it/s] Loading weights: 100%\|██████████\| 760/760 [00:06<00:00, 117.59it/s]
	Installed CUDA version 12.4 does not match the version torch was compiled with 12.8 but since the APIs are compatible, accepting this combination