Instructions to use LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0 with Transformers:

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

pipe = pipeline("text-generation", model="LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0")
model = AutoModelForCausalLM.from_pretrained("LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0", device_map="auto")
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 Settings

vLLM

How to use LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0

SGLang

How to use LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0 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 "LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0" \
    --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": "LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0",
		"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 "LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0" \
        --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": "LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0 with Docker Model Runner:
```
docker model run hf.co/LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Starfallen Snow Fantasy 24B 3.2

If there was snow that would fall forever.... Can I hide this feeling that continues towards you?~

Song: Changin’ My Life - Eternal Snow

Listen Here: https://www.youtube.com/watch?v=ZIF1YRz9N7E

So.... I'm kinda back, I hope. This was my attempt at trying to get a stellar like model out of Mistral 3.2 24b, I think I got most of it down besides a few quirks. It's not quite what I want to make in the future, but it's got good vibes. I like it, so try please?

Chat Template: Mistral V7 or ChatML

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the Linear DELLA merge method using ./Mistral-Small-3.2-24B-Instruct-2506-ChatML as a base.

Models Merged

The following models were included in the merge:

zerofata/MS3.2-PaintedFantasy-24B
Gryphe/Codex-24B-Small-3.2
Delta-Vector/MS3.2-Austral-Winton

Thanks to

Zerofata: for PaintedFantasy
Delta-Vector: for Austral Winton
Gryphe: for Codex

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Delta-Vector/MS3.2-Austral-Winton
    parameters:
      weight: 0.3
      density: 0.25
  - model: Gryphe/Codex-24B-Small-3.2
    parameters:
      weight: 0.1
      density: 0.4
  - model: zerofata/MS3.2-PaintedFantasy-24B
    parameters:
      weight: 0.4
      density: 0.5
merge_method: della_linear
base_model: anthracite-core/Mistral-Small-3.2-24B-Instruct-2506-ChatML
parameters:
  epsilon: 0.05
  lambda: 1
merge_method: della_linear
dtype: bfloat16
tokenizer:
  source: ./Codex-24B-Small-3.2

Downloads last month: 60

Safetensors

Model size

24B params

Tensor type

BF16

Model tree for LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0

Quantizations

5 models

Paper for LyraNovaHeart/Starfallen-Snow-Fantasy-24B-MS3.2-v0.0

DELLA-Merging: Reducing Interference in Model Merging through Magnitude-Based Sampling

Paper • 2406.11617 • Published Jun 17, 2024 • 12