Instructions to use Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound with Transformers:

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

pipe = pipeline("image-text-to-text", model="Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound")
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("Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound")
model = AutoModelForMultimodalLM.from_pretrained("Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound", 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 Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound",
		"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/Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound

SGLang

How to use Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound 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 "Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound" \
    --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": "Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound",
		"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 "Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound" \
        --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": "Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound",
		"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 Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound with Docker Model Runner:
```
docker model run hf.co/Pilcothink/Qwen3.5-9B-MixedInt4-AutoRound
```

Qwen3.5-9B-MixedInt4-AutoRound / quantization_config.json

Pilcothink

Upload Qwen3.5-9B MixedInt4 AutoRound

6b287f0 verified 8 days ago

Raw

History Blame Contribute Delete

5.27 kB

	{
	"bits": 4,
	"data_type": "int",
	"group_size": 32,
	"sym": true,
	"scale_dtype": "None",
	"autoround_version": "0.14.2",
	"block_name_to_quantize": "model.language_model.layers",
	"quant_method": "auto-round",
	"packing_format": "auto_round:auto_gptq",
	"extra_config": {
	"model.language_model.layers.0.linear_attn.in_proj_qkv": {
	"bits": 8
	},
	"model.language_model.layers.0.linear_attn.in_proj_z": {
	"bits": 8
	},
	"model.language_model.layers.0.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.0.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.1.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.1.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.2.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.2.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.4.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.4.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.5.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.5.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.6.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.6.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.8.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.8.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.9.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.9.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.10.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.10.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.12.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.12.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.13.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.13.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.14.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.14.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.16.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.16.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.17.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.17.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.18.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.18.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.20.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.20.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.21.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.21.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.22.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.22.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.24.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.24.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.25.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.25.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.26.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.26.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.28.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.28.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.29.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.29.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.30.linear_attn.in_proj_b": {
	"bits": 8
	},
	"model.language_model.layers.30.linear_attn.in_proj_a": {
	"bits": 8
	},
	"model.language_model.layers.0.linear_attn.out_proj": {
	"bits": 8
	},
	"model.language_model.layers.1.linear_attn.out_proj": {
	"bits": 8
	},
	"model.language_model.layers.2.linear_attn.out_proj": {
	"bits": 8
	},
	"model.language_model.layers.2.mlp.down_proj": {
	"bits": 8
	},
	"model.language_model.layers.3.self_attn.o_proj": {
	"bits": 8
	},
	"model.language_model.layers.7.self_attn.o_proj": {
	"bits": 8
	},
	"model.language_model.layers.11.self_attn.o_proj": {
	"bits": 8
	}
	}
	}