Instructions to use FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw with Transformers:

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

pipe = pipeline("text-generation", model="FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw")
model = AutoModelForCausalLM.from_pretrained("FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw", 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 FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw

SGLang

How to use FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw 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 "FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw" \
    --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": "FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw",
		"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 "FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw" \
        --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": "FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw with Docker Model Runner:
```
docker model run hf.co/FluffyKaeloky/L3.3-70B-Euryale-v2.3-exl2-4.5bpw
```

Many thanks. Sorry to bother, but it is possible to do a 6.0 bpw one?

by Panchovix - opened Dec 10, 2024

Discussion

Panchovix

Dec 10, 2024

Many thanks for the quant. I was wondering, if it was possible to do a quant at 6BPW. Thanks in advance!

SporkySporkness

Dec 10, 2024

And I'd happily take a 5bpw! Thanks :)

FluffyKaeloky

Owner Dec 10, 2024

Sure thing, I'll launch the quants and upload them as soon as I can.

FluffyKaeloky

Owner Dec 11, 2024

I have uploaded a 6bpw version. As I was about to upload the 5.0bpw, I noticed Dracones uploaded a whole range of exl2 quants. Feel free to download his for 5.0bpw :)

Panchovix changed discussion status to closed Dec 11, 2024

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