Instructions to use Vortex5/NoctyxCosma-12B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Vortex5/NoctyxCosma-12B with Transformers:

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

pipe = pipeline("text-generation", model="Vortex5/NoctyxCosma-12B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Vortex5/NoctyxCosma-12B")
model = AutoModelForCausalLM.from_pretrained("Vortex5/NoctyxCosma-12B", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Vortex5/NoctyxCosma-12B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Vortex5/NoctyxCosma-12B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Vortex5/NoctyxCosma-12B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Vortex5/NoctyxCosma-12B

SGLang

How to use Vortex5/NoctyxCosma-12B 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 "Vortex5/NoctyxCosma-12B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Vortex5/NoctyxCosma-12B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "Vortex5/NoctyxCosma-12B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Vortex5/NoctyxCosma-12B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Vortex5/NoctyxCosma-12B with Docker Model Runner:
```
docker model run hf.co/Vortex5/NoctyxCosma-12B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

NoctyxCosma-12B

Overview

NoctyxCosma-12B was created by merging Harmony-Bird-12B, Chrysologus-12B, Rei-V3-KTO-12B, Scarlet-Eclipse-12B, Lunar-Nexus-12B, and Red-Synthesis-12B using a custom merge method.

Show merge configuration

models:
  - model: Vortex5/Harmony-Bird-12B
  - model: Retreatcost/Chrysologus-12B
  - model: Delta-Vector/Rei-V3-KTO-12B
  - model: Vortex5/Scarlet-Eclipse-12B
  - model: Vortex5/Lunar-Nexus-12B
  - model: Vortex5/Red-Synthesis-12B
merge_method: smi_oni
parameters:
  k_core: 2
  strength_core: 1.0
  consensus_core: 0.48
  drop_cos: 0.10
  drop_min: 3
  strength_nov: 0.45
  novelty_budget: 0.22
  consensus_nov: 0.32
  conflict_bonus: 0.18
dtype: bfloat16
tokenizer:
  source: Vortex5/Red-Synthesis-12B