Instructions to use cyankiwi/Nex-N2-mini-AWQ-INT4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cyankiwi/Nex-N2-mini-AWQ-INT4 with Transformers:

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

pipe = pipeline("text-generation", model="cyankiwi/Nex-N2-mini-AWQ-INT4")
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("cyankiwi/Nex-N2-mini-AWQ-INT4")
model = AutoModelForMultimodalLM.from_pretrained("cyankiwi/Nex-N2-mini-AWQ-INT4", 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 cyankiwi/Nex-N2-mini-AWQ-INT4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cyankiwi/Nex-N2-mini-AWQ-INT4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "cyankiwi/Nex-N2-mini-AWQ-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cyankiwi/Nex-N2-mini-AWQ-INT4

SGLang

How to use cyankiwi/Nex-N2-mini-AWQ-INT4 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 "cyankiwi/Nex-N2-mini-AWQ-INT4" \
    --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": "cyankiwi/Nex-N2-mini-AWQ-INT4",
		"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 "cyankiwi/Nex-N2-mini-AWQ-INT4" \
        --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": "cyankiwi/Nex-N2-mini-AWQ-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cyankiwi/Nex-N2-mini-AWQ-INT4 with Docker Model Runner:
```
docker model run hf.co/cyankiwi/Nex-N2-mini-AWQ-INT4
```

error trying to run mini on a single 5090

by robert896r1 - opened Jun 8

Discussion

robert896r1

Jun 8

[2026-06-08 21:40:14] Init torch distributed begin.
[2026-06-08 21:40:14] Init torch distributed ends. elapsed=0.16 s, mem usage=0.11 GB
[2026-06-08 21:40:15] Load weight begin. avail mem=30.06 GB
[2026-06-08 21:40:15] Multimodal attention backend not set. Use triton_attn.
[2026-06-08 21:40:15] Using triton_attn as multimodal attention backend.
[2026-06-08 21:40:15] Acceleration for non-quantized schemes is not supported by Compressed Tensors. Falling back to UnquantizedLinearMethod
[transformers] torch_dtype is deprecated! Use dtype instead!
[2026-06-08 21:40:15] Scheduler hit an exception: Traceback (most recent call last):
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 4044, in run_scheduler_process
scheduler = Scheduler(
^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 437, in init
self.init_model_worker()
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 718, in init_model_worker
self.init_tp_model_worker()
File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 673, in init_tp_model_worker
self.tp_worker = TpModelWorker(**worker_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/managers/tp_worker.py", line 262, in init
self._init_model_runner()
File "/sgl-workspace/sglang/python/sglang/srt/managers/tp_worker.py", line 347, in _init_model_runner
self._model_runner = ModelRunner(
^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 535, in init
self.initialize(pre_model_load_memory)
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 655, in initialize
self.load_model()
File "/sgl-workspace/sglang/python/sglang/srt/model_executor/model_runner.py", line 1465, in load_model
self.model = self.loader.load_model(
^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_loader/loader.py", line 718, in load_model
model = _initialize_model(
^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/model_loader/loader.py", line 292, in _initialize_model
return model_class(**kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/qwen3_5.py", line 1608, in init
super().init(config, quant_config, prefix, language_model_cls)
File "/sgl-workspace/sglang/python/sglang/srt/models/qwen3_vl.py", line 1124, in init
self.model = language_model_cls(
^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/qwen3_5.py", line 1237, in init
super().init(config=config, quant_config=quant_config, prefix=prefix)
File "/sgl-workspace/sglang/python/sglang/srt/models/qwen3_5.py", line 1046, in init
self.layers, self._start_layer, self._end_layer = make_layers(
^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/utils/common.py", line 696, in make_layers
+ get_offloader().wrap_modules(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/utils/offloader.py", line 36, in wrap_modules
return list(all_modules_generator)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/utils/common.py", line 698, in
layer_fn(idx=idx, prefix=add_prefix(idx, prefix))
File "/sgl-workspace/sglang/python/sglang/srt/models/qwen3_5.py", line 1037, in get_layer
return layer_class(
^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/qwen3_5.py", line 554, in init
self.mlp = Qwen2MoeSparseMoeBlock(
^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/models/qwen2_moe.py", line 255, in init
self.experts = get_moe_impl_class(quant_config)(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/layers/moe/fused_moe_triton/layer.py", line 285, in init
self.quant_method = quant_config.get_quant_method(self, prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py", line 180, in get_quant_method
layer.scheme = self.get_moe_scheme(layer=layer, layer_name=prefix)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py", line 709, in get_moe_scheme
elif self._is_dynamic_token_w8a8(weight_quant, input_quant):
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/sgl-workspace/sglang/python/sglang/srt/layers/quantization/compressed_tensors/compressed_tensors.py", line 389, in _is_dynamic_token_w8a8
is_8_bits = weight_quant.num_bits == input_quant.num_bits == 8
^^^^^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'num_bits'

cpatonn

cyankiwi org Jun 9

Thank you for using cyankiwi models. Currently for this model, please use it with vllm :)

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment