Instructions to use Delta-Vector/MS3.2-Austral-Winton with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Delta-Vector/MS3.2-Austral-Winton with Transformers:

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

pipe = pipeline("text-generation", model="Delta-Vector/MS3.2-Austral-Winton")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Delta-Vector/MS3.2-Austral-Winton")
model = AutoModelForCausalLM.from_pretrained("Delta-Vector/MS3.2-Austral-Winton")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Delta-Vector/MS3.2-Austral-Winton with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Delta-Vector/MS3.2-Austral-Winton"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Delta-Vector/MS3.2-Austral-Winton",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Delta-Vector/MS3.2-Austral-Winton

SGLang

How to use Delta-Vector/MS3.2-Austral-Winton 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 "Delta-Vector/MS3.2-Austral-Winton" \
    --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": "Delta-Vector/MS3.2-Austral-Winton",
		"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 "Delta-Vector/MS3.2-Austral-Winton" \
        --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": "Delta-Vector/MS3.2-Austral-Winton",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Delta-Vector/MS3.2-Austral-Winton with Docker Model Runner:
```
docker model run hf.co/Delta-Vector/MS3.2-Austral-Winton
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Trained by Delta-Vector

Overview

Austral 24B - Winton

Codex Finetune Mistral Based KTO enhanced Adventure/Roleplay generalist 24B Sized model

More than 1.5-metres tall, about six-metres long and up to 1000-kilograms heavy, Australovenator Wintonensis was a fast and agile hunter. The largest known Australian theropod.

This is a finetune of Codex 24B to be a generalist Roleplay/Adventure model. I've removed some of the "slops" that i noticed in an otherwise great model aswell as improving the general writing of the model, This was a multi-stage finetune, all previous checkpoints are released aswell. In testing it has shown to be a great model for Adventure cards & Roleplay, Often pushing the plot forward better then other models, While avoiding some of the slops you'd find in models from Drummer and Co.

Support my finetunes / Me on Kofi: https://Ko-fi.com/deltavector | Thank you to Auri for helping/Testing ♥

Quants

Quants Formats

GGUFFor use with LLama.cpp & Forks (coming soon!))
EXL3For use with TabbyAPI (Coming soon!)

Chat Format

This model utilizes ChatML.

<|im_start|>user
Hi there!<|im_end|>
<|im_start|>assistant
Nice to meet you!<|im_end|>
<|im_start|>user
Can I ask a question?<|im_end|>
<|im_start|>assistant

Training

As the the Austral/Francois tradition, I built off another great finetune Codex-24B, I did 4 epochs ontop with roughly the same datamix as Francois-Huali/Austral 70B as a R128 Lora, then KTO alignment with a mix of Instruct/Small writing datasets and then finally another 4 epoch SFT with Rep_remover (Thanks Pocket!)

Config(Post-KTO SFT)

https://wandb.ai/new-eden/austral/runs/i85da0c6?nw=nwuserdeltavector

This model was trained over 4 epochs using 8 x A100s (Ty to my work, Cognitive Computations) for the base SFT, Then i used KTO to clean up some coherency issues for 1 epoch, then finally training for another 4 epochs on Rep_Remover to delete slops. Total was roughly 80 hours total.