Instructions to use oxyapi/oxy-1-small-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use oxyapi/oxy-1-small-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="oxyapi/oxy-1-small-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("oxyapi/oxy-1-small-GGUF", dtype="auto")

llama-cpp-python

How to use oxyapi/oxy-1-small-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="oxyapi/oxy-1-small-GGUF",
	filename="oxy-1-small.F16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use oxyapi/oxy-1-small-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf oxyapi/oxy-1-small-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf oxyapi/oxy-1-small-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf oxyapi/oxy-1-small-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf oxyapi/oxy-1-small-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf oxyapi/oxy-1-small-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf oxyapi/oxy-1-small-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf oxyapi/oxy-1-small-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf oxyapi/oxy-1-small-GGUF:Q4_K_M

Use Docker

docker model run hf.co/oxyapi/oxy-1-small-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use oxyapi/oxy-1-small-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "oxyapi/oxy-1-small-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oxyapi/oxy-1-small-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/oxyapi/oxy-1-small-GGUF:Q4_K_M

SGLang

How to use oxyapi/oxy-1-small-GGUF 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 "oxyapi/oxy-1-small-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oxyapi/oxy-1-small-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "oxyapi/oxy-1-small-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oxyapi/oxy-1-small-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use oxyapi/oxy-1-small-GGUF with Ollama:
```
ollama run hf.co/oxyapi/oxy-1-small-GGUF:Q4_K_M
```

Unsloth Studio new

How to use oxyapi/oxy-1-small-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for oxyapi/oxy-1-small-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for oxyapi/oxy-1-small-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for oxyapi/oxy-1-small-GGUF to start chatting

Docker Model Runner
How to use oxyapi/oxy-1-small-GGUF with Docker Model Runner:
```
docker model run hf.co/oxyapi/oxy-1-small-GGUF:Q4_K_M
```

Lemonade

How to use oxyapi/oxy-1-small-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull oxyapi/oxy-1-small-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.oxy-1-small-GGUF-Q4_K_M

List all available models

lemonade list

Repetition issues

by deleted - opened Dec 1, 2024

Discussion

deleted

Dec 1, 2024

Sadly it seems this model has a bad habit of repeating itself, it happened very quickly during my RPs with it. It also seems not as smart as NeMo but maybe I need to test it some more... Apart from the repetition issue, it's not bad.

TornadoAI

Oxygen org Dec 1, 2024

The quality obviously depends on the GGUF you have chosen, with the base model I did not encounter any repetition problems, have you tried to adjust the repetition_penalty? or use another GGUF? what can I do on my side to spot this problem?

deleted

Dec 2, 2024

•

edited Dec 2, 2024

I used a Q5KM from another person on here, and no repetition penalty or anything else as I don't really require that usually with bigger models... I'm testing again now to try and reproduce it.

It also seems to really like NOT following instructions 😅

TornadoAI

Oxygen org Dec 2, 2024

Oh sorry for that, I never use GGUF I didn't test anything before uploading.. I can understand that you don't like the model there is no problem don't force yourself to use it if it doesn't seem to fit!

deleted

Dec 2, 2024

It's a promising model, I just wanted to help find what's wrong 🙁

TornadoAI

Oxygen org Dec 2, 2024

Two other people on reddit talked about this problem, I wonder if it comes from a quants or if it's all of them that are bugged.. have you had any problems with the bartowski or mradermacher quants?

deleted

Dec 2, 2024

Yeah I started with mrademacher then yours and had the same issues

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