Instructions to use Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw with Transformers:

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

pipe = pipeline("text-generation", model="Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw")
model = AutoModelForCausalLM.from_pretrained("Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw

SGLang

How to use Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw 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 "Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw" \
    --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": "Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw",
		"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 "Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw" \
        --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": "Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw with Docker Model Runner:
```
docker model run hf.co/Jellon/Qwen2.5-14b-Kunou-v1-exl2-4bpw
```

Jellon commited on Dec 22, 2024

Commit

fdca068

verified ·

1 Parent(s): dfdd984

Update README.md

Browse files

Files changed (1) hide show

README.md +4 -1

README.md CHANGED Viewed

@@ -3,7 +3,7 @@ library_name: transformers
 license: other
 license_name: qwen
 license_link: https://huggingface.co/Qwen/Qwen2.5-14B-Instruct/blob/main/LICENSE
-base_model: Qwen/Qwen2.5-14B-Instruct
 tags:
 - generated_from_trainer
 model-index:
@@ -11,6 +11,9 @@ model-index:
   results: []
 ---
 ![Kunou](https://huggingface.co/Sao10K/72B-Qwen2.5-Kunou-v1/resolve/main/knn.png)
 **Elder Sister Versions for Heavier Use!**

 license: other
 license_name: qwen
 license_link: https://huggingface.co/Qwen/Qwen2.5-14B-Instruct/blob/main/LICENSE
+base_model: Sao10K/14B-Qwen2.5-Kunou-v1
 tags:
 - generated_from_trainer
 model-index:
   results: []
 ---
+4bpw exl2 quant of: https://huggingface.co/Sao10K/14B-Qwen2.5-Kunou-v1
+---
 ![Kunou](https://huggingface.co/Sao10K/72B-Qwen2.5-Kunou-v1/resolve/main/knn.png)
 **Elder Sister Versions for Heavier Use!**