Instructions to use 24B-Suite/Mergedonia-PROMETHEUS-24B-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use 24B-Suite/Mergedonia-PROMETHEUS-24B-v1 with Transformers:

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

pipe = pipeline("text-generation", model="24B-Suite/Mergedonia-PROMETHEUS-24B-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("24B-Suite/Mergedonia-PROMETHEUS-24B-v1")
model = AutoModelForCausalLM.from_pretrained("24B-Suite/Mergedonia-PROMETHEUS-24B-v1")
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 24B-Suite/Mergedonia-PROMETHEUS-24B-v1 with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/24B-Suite/Mergedonia-PROMETHEUS-24B-v1

SGLang

How to use 24B-Suite/Mergedonia-PROMETHEUS-24B-v1 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 "24B-Suite/Mergedonia-PROMETHEUS-24B-v1" \
    --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": "24B-Suite/Mergedonia-PROMETHEUS-24B-v1",
		"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 "24B-Suite/Mergedonia-PROMETHEUS-24B-v1" \
        --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": "24B-Suite/Mergedonia-PROMETHEUS-24B-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use 24B-Suite/Mergedonia-PROMETHEUS-24B-v1 with Docker Model Runner:
```
docker model run hf.co/24B-Suite/Mergedonia-PROMETHEUS-24B-v1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

PROMETHEUS (VERSION P): STABILIZED

🌄 Mergedonia PROMETHEUS 24B v1

A highly experimental new merge method which combines several official and custom methods into a complex 6-stage pipeline.

architecture: MistralForCausalLM
merge_method: prometheus # Version P
base_model: B:/24B/Darkhn--Magistral-2509-24B-Text-Only
parameters:
  # --- CABS + RAM Core Settings ---
  default_n_m_ratio: "1/4" 
  ram_r: 0.1
  ram_alpha: 0.2
  della_eps: 0.05
  gamma: 0.001
  breadcrumbs_density: 0.9
  select_topk: 0.5     
  # --- Advanced Toggles ---
  use_sign_consensus: true 
  della_row_wise: true
  sce_weighting: true
  lambda: 1.0
  auto_pad: true
  skip_on_mismatch: true
  # --- Quantum Frankenstein Glue + Shannon Entropy ---
  iota: 0.1
  zeta: 16
  chi: 0.15
  gamma: 1.0
  # --- AETHERIC SYMPLECTIC GROUNDING ---
  asg_resonance: 0.3
  asg_grounding: 0.7
  # --- FLUX Frankenstein Surgery (QUICK TEST PLATINUM) ---
  frankenstein_surgery: true
  phi_optim: true
  phi: 0.4
  rho: 0.15
  mu: 0.5
  tau: 2.5
  alpha: 0.5
  # --- Hilbert Transform Slingshot ---
  eta: 1.2              
  delta: 0.915
  mercury_retrograde_gravity: 1.0
  pi: 3141
  tol: 1.0e-9
  max_iter: 100
  kappa: 0.8
models:
  # --- TIER 1 ---
  - model: B:/24B/!models--TheDrummer--Precog-24B-v1
    parameters:
      weight: 1.0
      n_m_ratio: "1/5" ## Balanced for Mergedonia ## "1/2" # 50% density. Gets first pick and claims a massive footprint.
      allow_sce_steal: true
      fill_remaining: false
  # --- TIER 2 ---
  - model: B:/24B/TheDrummer--Magidonia-24B-v4.3
    parameters:
      weight: 1.0
      n_m_ratio: "1/4" ## 25% density. Claims a fourth of whatever Precog left behind. ##
      allow_sce_steal: true
      fill_remaining: false
  # --- TIER 3 ---
  - model: B:/24B/TheDrummer--Cydonia-24B-v4.3
    parameters:
      weight: 1.0
      n_m_ratio: "1/3" # No override: Falls back to global default[1, 4] (25% density)
      allow_sce_steal: true
      fill_remaining: false
  - model: B:/24B/!BeaverAI_Fallen-Mistral-Small-3.1-24B-v1e_textonly
    parameters:
      weight: 1.0
      n_m_ratio: "1/2" # No override: Falls back to global default [1, 4] (25% density)
      allow_sce_steal: true
      fill_remaining: false
  - model: B:/24B/!models--TheDrummer--Rivermind-24B-v1
    parameters:
      weight: 1.0
      n_m_ratio: "1/2" ## We use an increasing n_m_ratio for balanced Mergedonia, but normally a decreasing ratio produces the desired CABS model "hierarchy" ##
      # No override: Falls back to global default [1, 4] (25% density)
      allow_sce_steal: true # false # STOP THE STEAL: Rivermind can only have leftovers
      fill_remaining: true
tokenizer:
  source: union
chat_template: auto
dtype: float32
out_dtype: bfloat16
name: 🌄 Mergedonia-Prometheus-24B-v1