Instructions to use timdettmers/guanaco-65b-merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use timdettmers/guanaco-65b-merged with Transformers:

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

pipe = pipeline("text-generation", model="timdettmers/guanaco-65b-merged")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("timdettmers/guanaco-65b-merged")
model = AutoModelForCausalLM.from_pretrained("timdettmers/guanaco-65b-merged", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use timdettmers/guanaco-65b-merged with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "timdettmers/guanaco-65b-merged"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "timdettmers/guanaco-65b-merged",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/timdettmers/guanaco-65b-merged

SGLang

How to use timdettmers/guanaco-65b-merged 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 "timdettmers/guanaco-65b-merged" \
    --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": "timdettmers/guanaco-65b-merged",
		"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 "timdettmers/guanaco-65b-merged" \
        --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": "timdettmers/guanaco-65b-merged",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use timdettmers/guanaco-65b-merged with Docker Model Runner:
```
docker model run hf.co/timdettmers/guanaco-65b-merged
```

how "merged" is this one?

by sirus - opened Jun 12, 2023

Discussion

sirus

Jun 12, 2023

I've seen some pretty crazy lora merges. Just wondering the details of this model. It's currently number 2 on the open llm leaderboard, but the model card is blank.

So mysterious

sirus

Jun 13, 2023

also this appears not to be ggml

simsim314

Jun 16, 2023

It's just part of qlora research, you can lookup in other timdettmers models. Usually no one starts from 65B model, and it's pretty tedious to explain the same thing in dozens of similar model cards. Most people here are already familiar with how to use it, if you are not - check the qlora github repository mantioned above, there is a lot of good explanations there.

Yes it's not ggml - it uses the standard huggingface transformer library. Check out this colab for inference example.

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