Instructions to use cyankiwi/North-Mini-Code-1.0-AWQ-INT4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cyankiwi/North-Mini-Code-1.0-AWQ-INT4 with Transformers:

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

pipe = pipeline("text-generation", model="cyankiwi/North-Mini-Code-1.0-AWQ-INT4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("cyankiwi/North-Mini-Code-1.0-AWQ-INT4")
model = AutoModelForCausalLM.from_pretrained("cyankiwi/North-Mini-Code-1.0-AWQ-INT4", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cyankiwi/North-Mini-Code-1.0-AWQ-INT4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cyankiwi/North-Mini-Code-1.0-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/North-Mini-Code-1.0-AWQ-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cyankiwi/North-Mini-Code-1.0-AWQ-INT4

SGLang

How to use cyankiwi/North-Mini-Code-1.0-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/North-Mini-Code-1.0-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/North-Mini-Code-1.0-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/North-Mini-Code-1.0-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/North-Mini-Code-1.0-AWQ-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cyankiwi/North-Mini-Code-1.0-AWQ-INT4 with Docker Model Runner:
```
docker model run hf.co/cyankiwi/North-Mini-Code-1.0-AWQ-INT4
```

"Only symmetric quantization is supported for MoE"

by ydawei - opened Jun 11

Discussion

ydawei

Jun 11

Got this error, on vLLM 0.22.1:

EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/cohere2_moe.py", line 562, in __init__
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     self.model = Cohere2MoeModel(
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]                  ^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/compilation/decorators.py", line 383, in __init__
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     old_init(self, *args, **kwargs)
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/cohere2_moe.py", line 402, in __init__
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     self.start_layer, self.end_layer, self.layers = make_layers(
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]                                                     ^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/utils.py", line 643, in make_layers
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     + get_offloader().wrap_modules(
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/offloader/base.py", line 104, in wrap_modules
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     return list(modules_generator)
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]            ^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/utils.py", line 644, in <genexpr>
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     layer_fn(prefix=f"{prefix}.{idx}") for idx in range(start_layer, end_layer)
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/cohere2_moe.py", line 404, in <lambda>
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     lambda prefix: Cohere2MoeDecoderLayer(
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]                    ^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/cohere2_moe.py", line 359, in __init__
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     self.mlp = Cohere2Moe(
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]                ^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/models/cohere2_moe.py", line 300, in __init__
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     self.experts = FusedMoE(
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]                    ^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/layers/fused_moe/layer.py", line 368, in __init__
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     self.quant_method: FusedMoEMethodBase = _get_quant_method()
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]                                             ^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/layers/fused_moe/layer.py", line 360, in _get_quant_method
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     quant_method = self.quant_config.get_quant_method(self, prefix)
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors.py", line 184, in get_quant_method
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     return CompressedTensorsMoEMethod.get_moe_method(
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe/compressed_tensors_moe.py", line 114, in get_moe_method
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     return CompressedTensorsWNA16MoEMethod(
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]   File "/usr/local/lib/python3.12/dist-packages/vllm/model_executor/layers/quantization/compressed_tensors/compressed_tensors_moe/compressed_tensors_moe_wna16.py", line 50, in __init__
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]     assert weight_quant.symmetric, (
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165]            ^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=83) ERROR 06-11 06:46:38 [core.py:1165] AssertionError: Only symmetric quantization is supported for MoE

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