Instructions to use KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b with Transformers:

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

pipe = pipeline("text-generation", model="KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b")
model = AutoModelForCausalLM.from_pretrained("KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b", 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 KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b

SGLang

How to use KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b 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 "KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b" \
    --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": "KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b",
		"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 "KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b" \
        --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": "KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b with Docker Model Runner:
```
docker model run hf.co/KaraKaraWitch/LLENN-v0.69420-Qwen2.5-72b
```

KaraKaraWitch commited on Oct 30, 2024

Commit

3162216

verified ·

1 Parent(s): 80bdfa1

Update README.md

Browse files

Files changed (1) hide show

README.md +22 -0

README.md CHANGED Viewed

@@ -55,4 +55,26 @@ base_model: Qwen/Qwen2.5-72B
 parameters:
   normalize: true
 dtype: bfloat16
 ```

 parameters:
   normalize: true
 dtype: bfloat16
+```
+### Prompt Format
+ChatML works for the most part.
+### Sampler Settings
+Personally I use the following:
+```
+Temp: 1.2
+Min P: 0.07
+Rep Pen: 1.1
+```
+Others have suggested the following:
+```
+Temp: 1.1
+Top P: 0.98
+Min P: 0.05
 ```