Not-For-All-Audiences

Instructions to use Kaoeiri/MS-MagpantheonselRP-22B-13.0 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Kaoeiri/MS-MagpantheonselRP-22B-13.0 with Transformers:

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

pipe = pipeline("text-generation", model="Kaoeiri/MS-MagpantheonselRP-22B-13.0")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Kaoeiri/MS-MagpantheonselRP-22B-13.0")
model = AutoModelForCausalLM.from_pretrained("Kaoeiri/MS-MagpantheonselRP-22B-13.0", 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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Kaoeiri/MS-MagpantheonselRP-22B-13.0 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Kaoeiri/MS-MagpantheonselRP-22B-13.0"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Kaoeiri/MS-MagpantheonselRP-22B-13.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Kaoeiri/MS-MagpantheonselRP-22B-13.0

SGLang

How to use Kaoeiri/MS-MagpantheonselRP-22B-13.0 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 "Kaoeiri/MS-MagpantheonselRP-22B-13.0" \
    --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": "Kaoeiri/MS-MagpantheonselRP-22B-13.0",
		"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 "Kaoeiri/MS-MagpantheonselRP-22B-13.0" \
        --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": "Kaoeiri/MS-MagpantheonselRP-22B-13.0",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Kaoeiri/MS-MagpantheonselRP-22B-13.0 with Docker Model Runner:
```
docker model run hf.co/Kaoeiri/MS-MagpantheonselRP-22B-13.0
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

merge

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using unsloth/Mistral-Small-Instruct-2409 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  # Core Fiction and Character Detail Models
  - model: Kaoeiri/MS_Moingooistral-2409-22B
    parameters:
      weight: 0.36  # Balanced for nuanced character and story details
      density: 1.15 # Slightly reduced for better interaction with other models

  - model: Kaoeiri/MS-Magpantheonsel-lark-v4x1.6.2-Cydonia-vXXX-22B-5
    parameters:
      weight: 1.0   # Primary engine remains untouched
      density: 0.85 # Retained for depth and coherence

  # World Building & Character Interaction
  - model: Kaoeiri/MS-Inky-2409-22B
    parameters:
      weight: 0.38  # Balanced to allow subtle world-building
      density: 0.78 # Reduced slightly to integrate smoother interaction

  - model: Gryphe/Pantheon-RP-Pure-1.6.2-22b-Small
    parameters:
      weight: 0.42  # Optimized for character interaction
      density: 0.80 # Elevated slightly for smoother roleplay context

  # Character Development Core
  - model: DigitalSouls/BlackSheep-DigitalSoul-22B
    parameters:
      weight: 0.30  # Balanced for character dynamics
      density: 0.78 # Slightly enhanced for deeper conflicts and resolutions

  # Magical Elements
  - model: InferenceIllusionist/SorcererLM-22B
    parameters:
      weight: 0.14  # Retained for magical character depth
      density: 0.76 # Balanced to prevent overpowering influence

  - model: Envoid/Mistral-Small-NovusKyver
    parameters:
      weight: 0.14  # Balanced synergy with SorcererLM
      density: 0.76 # Synchronized with magical elements

  # Secondary Character Enhancement
  - model: TheDrummer/Cydonia-22B-v1.1
    parameters:
      weight: 0.16  # Balanced for supporting depth
      density: 0.70 # Slightly enhanced for richer secondary character narratives
  
  - model: crestf411/MS-sunfall-v0.7.0
    parameters:
      weight: 0.17  # Fine-tuned for precision
      density: 0.72 # Optimized for secondary character arcs

  - model: Kaoeiri/MS_a-coolyte-2409-22B
    parameters:
      weight: 0.16  # Reduced to reflect 1024-rank finetuning
      density: 0.65 # Limited to prevent excessive influence

  # Enhanced Personality Dynamics
  - model: Kaoeiri/MS_fujin-2409-22B
    parameters:
      weight: 0.09  # Introduced to replace Quadrosiac subtly
      density: 0.65 # Contributes to nuanced personality traits
  
  - model: Kaoeiri/MS_dampf-2409-22B
    parameters:
      weight: 0.12  # Enhances character complexity
      density: 0.70 # Integrated for personality interaction

  # Enhanced Story and Character Building
  - model: hf-100/Mistral-Small-Spellbound-StoryWriter-22B-instruct-0.2-chkpt-200-16-bit
    parameters:
      weight: 0.25  # Slightly enhanced for narrative engagement
      density: 0.75 # Elevated for richer storytelling

  - model: ArliAI/Mistral-Small-22B-ArliAI-RPMax-v1.1
    parameters:
      weight: 0.16  # Balanced for dynamic roleplay
      density: 0.68 # Elevated for smoother character interactions
  
  - model: Darkknight535/MS-Moonlight-22B-v3
    parameters:
      weight: 0.25  # Slightly enhanced for detail
      density: 0.68 # Adjusted for depth

  - model: concedo/Beepo-22B
    parameters:
      weight: 0.35  # Refined to prevent overpowering other layers
      density: 0.50 # Contributes subtle narrative shifts
  
  # Cultural and Character Depth
  - model: Saxo/Linkbricks-Horizon-AI-Japanese-Superb-V1-22B
    parameters:
      weight: 0.22  # Optimized for cultural enrichment
      density: 0.65 # Balanced for depth

  # Mermaid and Fantasy Personality
  - model: TroyDoesAI/BlackSheep-MermaidMistral-22B
    parameters:
      weight: 0.24  # Slightly enhanced for personality depth
      density: 0.74 # Elevated for immersive fantasy traits

merge_method: dare_ties
base_model: unsloth/Mistral-Small-Instruct-2409
parameters:
  density: 0.95    # Balanced for rich, cohesive character narratives
  epsilon: 0.035   # Optimized for smoother transitions
  lambda: 1.50     # Elevated for creative and coherent synergy
dtype: bfloat16
tokenizer_source: union