Instructions to use matchaaaaa/Chaifighter-20B-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use matchaaaaa/Chaifighter-20B-v2 with Transformers:

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

pipe = pipeline("text-generation", model="matchaaaaa/Chaifighter-20B-v2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("matchaaaaa/Chaifighter-20B-v2")
model = AutoModelForCausalLM.from_pretrained("matchaaaaa/Chaifighter-20B-v2")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use matchaaaaa/Chaifighter-20B-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "matchaaaaa/Chaifighter-20B-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "matchaaaaa/Chaifighter-20B-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/matchaaaaa/Chaifighter-20B-v2

SGLang

How to use matchaaaaa/Chaifighter-20B-v2 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 "matchaaaaa/Chaifighter-20B-v2" \
    --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": "matchaaaaa/Chaifighter-20B-v2",
		"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 "matchaaaaa/Chaifighter-20B-v2" \
        --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": "matchaaaaa/Chaifighter-20B-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use matchaaaaa/Chaifighter-20B-v2 with Docker Model Runner:
```
docker model run hf.co/matchaaaaa/Chaifighter-20B-v2
```

Any chance of increasing the context?

by DazzlingXeno - opened Jul 18, 2024

Discussion

DazzlingXeno

Jul 18, 2024

Would be great with 16k context or above.

matchaaaaa

Owner Jul 18, 2024

Hey @dazl1212,
I'm not sure if I can do 16K, but 8K might be possible now (RoPE the rest maybe?) with Fimbulvetr-11B-v2.1. I can whip that up for you with a few GGUFs if you like?
Thanks for trying my model! Have good day! :)

DazzlingXeno

Jul 19, 2024

•

edited Jul 19, 2024

Hi mate, that would be awesome. 8k with Chaifighter-20B-v2 will fit my needs great!

matchaaaaa

Owner Jul 23, 2024

Hey @dazl1212 ,
I'm uploading the tweaked model right now. In the meantime, please let me know if you need anything else. :)
Have a great day!

DazzlingXeno

Jul 23, 2024

Brilliant! Thank you!

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