Instructions to use cyankiwi/OmniCoder-9B-AWQ-BF16-INT4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cyankiwi/OmniCoder-9B-AWQ-BF16-INT4 with Transformers:

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

pipe = pipeline("text-generation", model="cyankiwi/OmniCoder-9B-AWQ-BF16-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)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("cyankiwi/OmniCoder-9B-AWQ-BF16-INT4")
model = AutoModelForMultimodalLM.from_pretrained("cyankiwi/OmniCoder-9B-AWQ-BF16-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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use cyankiwi/OmniCoder-9B-AWQ-BF16-INT4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cyankiwi/OmniCoder-9B-AWQ-BF16-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/OmniCoder-9B-AWQ-BF16-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cyankiwi/OmniCoder-9B-AWQ-BF16-INT4

SGLang

How to use cyankiwi/OmniCoder-9B-AWQ-BF16-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/OmniCoder-9B-AWQ-BF16-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/OmniCoder-9B-AWQ-BF16-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/OmniCoder-9B-AWQ-BF16-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/OmniCoder-9B-AWQ-BF16-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use cyankiwi/OmniCoder-9B-AWQ-BF16-INT4 with Docker Model Runner:
```
docker model run hf.co/cyankiwi/OmniCoder-9B-AWQ-BF16-INT4
```

OmniCoder-9B-AWQ-BF16-INT4

File size: 1,489 Bytes

12d607c

default_stage:
  default_modifiers:
    AWQModifier:
      config_groups:
        group_0:
          targets: [Linear]
          weights:
            num_bits: 4
            type: int
            symmetric: true
            group_size: 32
            strategy: group
            block_structure: null
            dynamic: false
            actorder: null
            scale_dtype: null
            zp_dtype: null
            observer: mse
            observer_kwargs: {}
          input_activations: null
          output_activations: null
          format: null
      targets: [Linear]
      ignore: ['re:.*embed_tokens', 're:.*linear_attn[.]conv1d', 're:.*linear_attn[.]in_proj_a',
        're:.*linear_attn[.]in_proj_b', 're:.*linear_attn.*', 're:model[.]visual.*', 're:mtp.*',
        lm_head]
      bypass_divisibility_checks: false
      mappings:
      - smooth_layer: re:model.*layers[.](3|7|11|15|19|23|27|31)[.]input_layernorm
        balance_layers: ['re:model.*layers[.](3|7|11|15|19|23|27|31)[.]self_attn[.]q_proj',
          're:model.*layers[.](3|7|11|15|19|23|27|31)[.]self_attn[.]k_proj', 're:model.*layers[.](3|7|11|15|19|23|27|31)[.]self_attn[.]v_proj']
        activation_hook_target: null
      - smooth_layer: re:model.*post_attention_layernorm
        balance_layers: ['re:model.*mlp[.]gate_proj', 're:model.*mlp[.]up_proj']
        activation_hook_target: null
      offload_device: !!python/object/apply:torch.device [cuda]
      duo_scaling: true
      n_grid: 20