Instructions to use bartowski/Llama-3.3-70B-Instruct-exl2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bartowski/Llama-3.3-70B-Instruct-exl2 with Transformers:

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

pipe = pipeline("text-generation", model="bartowski/Llama-3.3-70B-Instruct-exl2")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("bartowski/Llama-3.3-70B-Instruct-exl2", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use bartowski/Llama-3.3-70B-Instruct-exl2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bartowski/Llama-3.3-70B-Instruct-exl2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bartowski/Llama-3.3-70B-Instruct-exl2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/bartowski/Llama-3.3-70B-Instruct-exl2

SGLang

How to use bartowski/Llama-3.3-70B-Instruct-exl2 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 "bartowski/Llama-3.3-70B-Instruct-exl2" \
    --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": "bartowski/Llama-3.3-70B-Instruct-exl2",
		"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 "bartowski/Llama-3.3-70B-Instruct-exl2" \
        --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": "bartowski/Llama-3.3-70B-Instruct-exl2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use bartowski/Llama-3.3-70B-Instruct-exl2 with Docker Model Runner:
```
docker model run hf.co/bartowski/Llama-3.3-70B-Instruct-exl2
```

kaggle

by rakmik - opened Mar 23, 2025

Discussion

rakmik

Mar 23, 2025

bartowski/Llama-3.3-70B-Instruct-exl2
2.2 bits per weight

how to run it in kaggle on 2 gpu t4
Can you help me with the code to make it work?
Thank you.

rakmik

Mar 24, 2025

how to run bartowski/QwQ-32B-exl2 in colab t4

i want code like
python XXXXXXX.py -m "/content/QwQ-32B-exl2-3_0" -p "hi"

rakmik

Mar 24, 2025

https://github.com/kim90000/Xexl2/blob/main/XXexl2.ipynb

rakmik

Mar 24, 2025

!python examples/chat.py -m ../my_model2 -mode llama -cq4 -nfa -l 64

cash q4
Context length 64
without flash in colab t4
bartowski/QwQ-32B-exl2 Question: Can the model run in Colab T4? https://huggingface.co/bartowski/QwQ-32B-exl2/tree/3_0

Please edit the code

bartowski

Owner Mar 24, 2025

I don't think it would fit, 16GB won't be enough to load 3 bpw unless you use barely any context (which would be useless on a reasoning model)

rakmik

Mar 25, 2025

https://github.com/kim90000/suc-QwQ-32B-exl2-3_0/blob/main/suc_QwQ_32B_exl2_3_0%20(1).ipynb

Can you try the page and how can I make the topic useful and good especially for those who are only 16gb vram

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