Instructions to use LavaPlanet/Goliath120B-exl2_2-2.64bpw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LavaPlanet/Goliath120B-exl2_2-2.64bpw with Transformers:

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

pipe = pipeline("text-generation", model="LavaPlanet/Goliath120B-exl2_2-2.64bpw")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LavaPlanet/Goliath120B-exl2_2-2.64bpw")
model = AutoModelForCausalLM.from_pretrained("LavaPlanet/Goliath120B-exl2_2-2.64bpw", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use LavaPlanet/Goliath120B-exl2_2-2.64bpw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LavaPlanet/Goliath120B-exl2_2-2.64bpw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LavaPlanet/Goliath120B-exl2_2-2.64bpw",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/LavaPlanet/Goliath120B-exl2_2-2.64bpw

SGLang

How to use LavaPlanet/Goliath120B-exl2_2-2.64bpw 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 "LavaPlanet/Goliath120B-exl2_2-2.64bpw" \
    --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": "LavaPlanet/Goliath120B-exl2_2-2.64bpw",
		"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 "LavaPlanet/Goliath120B-exl2_2-2.64bpw" \
        --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": "LavaPlanet/Goliath120B-exl2_2-2.64bpw",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use LavaPlanet/Goliath120B-exl2_2-2.64bpw with Docker Model Runner:
```
docker model run hf.co/LavaPlanet/Goliath120B-exl2_2-2.64bpw
```

Another EXL2 version of AlpinDale's https://huggingface.co/alpindale/goliath-120b this one being at 2.64BPW and using the new experimental quant method of exllamav2.

Pippa llama2 Chat was used as the calibration dataset.

Can be run on two RTX 3090s w/ 24GB vram each.

Assuming Windows overhead, the following figures should be more or less close enough for estimation of your own use.

2.64BPW @ 4096 ctx
  Empty Ctx
    GPU Split:18/24
    GPU1: 19.8/24
    GPU2: 21.9/24
    10~ tk/s

Downloads last month: 14