Not-For-All-Audiences

nsfw

Instructions to use NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt with Transformers:

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

pipe = pipeline("text-generation", model="NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt")
model = AutoModelForMultimodalLM.from_pretrained("NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt")
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]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt

SGLang

How to use NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt 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 "NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt" \
    --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": "NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt",
		"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 "NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt" \
        --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": "NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt with Docker Model Runner:
```
docker model run hf.co/NeverSleep/Llama-3-Lumimaid-70B-v0.1-alt
```

Preferable to the normal version

by JoggyMuffin - opened May 8, 2024

Discussion

JoggyMuffin

May 8, 2024

For some reason the alt version seems much less prone to gpt-isms and getting stuck in a hole of generating the same-reply-but-with-different-words on multiple retries. Dialogue also seems more natural to the context/character and less synthetic sounding. Overall, I recommend people give this one a try if they aren't satisfied with non-alt version.

JoggyMuffin changed discussion title from Preferable to the nromal version to Preferable to the normal version May 8, 2024

BigHuggyD

May 9, 2024

For some reason the alt version seems much less prone to gpt-isms and getting stuck in a hole of generating the same-reply-but-with-different-words on multiple retries. Dialogue also seems more natural to the context/character and less synthetic sounding. Overall, I recommend people give this one a try if they aren't satisfied with non-alt version.

I'll have to give it a try. Was not a fan of 'non-alt'

Undi95

NeverSleep org May 9, 2024

It's because alt is epoch 2 and normal is epoch 1, Ikari liked e1 more and I liked e2 more, so we made the two available.
The OAS one is the alt one tho

Szarka

May 9, 2024

i prefer alt version too, but still lacking a bit. Compared to miqumaid merges

LLLLMAfficionado

May 17, 2024

•

edited May 17, 2024

miqumaid is very good, I haven't found anything that runs above how creative it can be, and naughty yeah, howver this one seems to produce more natural verbiage at times, not as overly formal or academic like miqumaid can be.

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