Instructions to use Fluxmire/Qwen-3.6-27B-Merlin with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Fluxmire/Qwen-3.6-27B-Merlin with Transformers:

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

pipe = pipeline("image-text-to-text", model="Fluxmire/Qwen-3.6-27B-Merlin")
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("Fluxmire/Qwen-3.6-27B-Merlin")
model = AutoModelForMultimodalLM.from_pretrained("Fluxmire/Qwen-3.6-27B-Merlin", 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 Fluxmire/Qwen-3.6-27B-Merlin with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Fluxmire/Qwen-3.6-27B-Merlin"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Fluxmire/Qwen-3.6-27B-Merlin",
		"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/Fluxmire/Qwen-3.6-27B-Merlin

SGLang

How to use Fluxmire/Qwen-3.6-27B-Merlin 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 "Fluxmire/Qwen-3.6-27B-Merlin" \
    --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": "Fluxmire/Qwen-3.6-27B-Merlin",
		"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 "Fluxmire/Qwen-3.6-27B-Merlin" \
        --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": "Fluxmire/Qwen-3.6-27B-Merlin",
		"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 Fluxmire/Qwen-3.6-27B-Merlin with Docker Model Runner:
```
docker model run hf.co/Fluxmire/Qwen-3.6-27B-Merlin
```

Qwen-3.6-27B-Merlin

A 27B vision-language checkpoint from the Qwen3.6-27B family, distributed as FP8 weights that load directly in vLLM. The upstream models this repository derives from are listed in NOTICE.

Checkpoint


Architecture	`Qwen3_5ForConditionalGeneration` — 64 decoder layers (hybrid linear / full attention), hidden size 5120, 27-block vision encoder
Context length	262 144 tokens
Weights	language-model `Linear` layers in FP8 (`compressed-tensors`, dynamic per-token activation scales); vision tower, linear-attention blocks, `lm_head` and MTP head in bfloat16
Files	`model-00001-of-00002.safetensors`, `model-00002-of-00002.safetensors` (~31 GB), `model-mtp.safetensors` (optional multi-token-prediction head), tokenizer, processor and chat template
Quantization record	`recipe.yaml`

Tokenizer, processor configuration and chat template are those of the Qwen3.6 family and are unchanged.

Serving with vLLM

Requires vLLM ≥ 0.23 with compressed-tensors FP8 support.

vllm serve magicworld7/Qwen-3.6-27B-Merlin \
  --revision <commit> \
  --data-parallel-size 2 \
  --max-model-len 81920 \
  --gpu-memory-utilization 0.9 \
  --enable-prefix-caching --enable-chunked-prefill

The server exposes the OpenAI-compatible chat API; image inputs are passed as image_url parts of the user message. generation_config.json holds the family's default sampling parameters (temperature 1.0, top_p 0.95, top_k 20); adjust them per use case.

Pin a revision when reproducibility matters — the repository may receive further commits.

Loading with Transformers

from transformers import AutoProcessor, AutoModelForImageTextToText
m = "magicworld7/Qwen-3.6-27B-Merlin"
processor = AutoProcessor.from_pretrained(m)
model = AutoModelForImageTextToText.from_pretrained(m, device_map="auto")

A recent Transformers release with compressed-tensors installed is needed to read the FP8 weights.

License

Apache License 2.0. LICENSE contains the license text; NOTICE contains the change notice and upstream attribution required by Section 4 of the license. "Qwen" is a trademark of Alibaba Cloud and is used only to identify the model family.

Downloads last month: 34

Safetensors

Model size

27B params

Tensor type

BF16

F8_E4M3

Model tree for Fluxmire/Qwen-3.6-27B-Merlin

Base model

Tooony133/Qwen-3.6-27B-SkinnyPete

Quantized

computer-vision-ai-lab/Qwen-3.6-27B-AmberKite

Quantized

(4)

this model