Instructions to use TheBloke/Airoboros-L2-70B-2.1-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBloke/Airoboros-L2-70B-2.1-GPTQ with Transformers:

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

pipe = pipeline("text-generation", model="TheBloke/Airoboros-L2-70B-2.1-GPTQ")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TheBloke/Airoboros-L2-70B-2.1-GPTQ")
model = AutoModelForCausalLM.from_pretrained("TheBloke/Airoboros-L2-70B-2.1-GPTQ")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use TheBloke/Airoboros-L2-70B-2.1-GPTQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TheBloke/Airoboros-L2-70B-2.1-GPTQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBloke/Airoboros-L2-70B-2.1-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TheBloke/Airoboros-L2-70B-2.1-GPTQ

SGLang

How to use TheBloke/Airoboros-L2-70B-2.1-GPTQ 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 "TheBloke/Airoboros-L2-70B-2.1-GPTQ" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBloke/Airoboros-L2-70B-2.1-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "TheBloke/Airoboros-L2-70B-2.1-GPTQ" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBloke/Airoboros-L2-70B-2.1-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TheBloke/Airoboros-L2-70B-2.1-GPTQ with Docker Model Runner:
```
docker model run hf.co/TheBloke/Airoboros-L2-70B-2.1-GPTQ
```

Weights updated?

by krao - opened Aug 29, 2023

Discussion

krao

Aug 29, 2023

Same here (as for the GGUF weights): looks like the weights have been updated:
https://huggingface.co/jondurbin/airoboros-l2-70b-2.1

Is this version based on the previous or current weights?

musicurgy

Aug 29, 2023

Looks like this is the "old" version. @TheBloke you should probably quantize the model again just to be safe, if it isn't too much work.

TheBloke

Owner Aug 29, 2023

New GPTQ weights will be up in 30-60 mins from now - nearly done

cgato

Aug 29, 2023

Gotta give him time to cook. :)

TheBloke

Owner Aug 29, 2023

•

edited Aug 29, 2023

They had to be done again (again) due to an issue in the base repo which caused my GPTQs to fail this morning, so ETA is currently around 2 hours

cgato

Aug 29, 2023

No problem! I've already used several of your quants. Way overdue, but I'm subbing to your Patreon. -.-

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