Instructions to use MixlyGames/Orpheus_KAISER with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MixlyGames/Orpheus_KAISER with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="MixlyGames/Orpheus_KAISER")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("MixlyGames/Orpheus_KAISER", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use MixlyGames/Orpheus_KAISER with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "MixlyGames/Orpheus_KAISER" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MixlyGames/Orpheus_KAISER", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/MixlyGames/Orpheus_KAISER
- SGLang
How to use MixlyGames/Orpheus_KAISER 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 "MixlyGames/Orpheus_KAISER" \ --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": "MixlyGames/Orpheus_KAISER", "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 "MixlyGames/Orpheus_KAISER" \ --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": "MixlyGames/Orpheus_KAISER", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use MixlyGames/Orpheus_KAISER with Docker Model Runner:
docker model run hf.co/MixlyGames/Orpheus_KAISER
| license: apache-2.0 | |
| language: | |
| - ru | |
| - en | |
| tags: | |
| - gpt | |
| - causal-lm | |
| - pytorch | |
| - transformers | |
| - Unsloth | |
| base_model: | |
| - MixlyGames/Orpheus_KAISER | |
| pipeline_tag: text-generation | |
| library_name: transformers | |
| # Orpheus-KAISER Model Card | |
| *Orpheus-KAISER** is a large and powerful Orpheus family model embodying the idea of absolute dominance and strength. | |
| The name refers to **Ungeziefer Kaiser** (The Roach Emperor) from Limbus Company, a transformed, overwhelming and extremely powerful form that symbolizes the pinnacle of power through radical transformation. | |
| ## Model Specifications | |
| - **Developer:** MixlyGames (OwlNestTeam) | |
| - **Model Type:** Causal Language Modeling | |
| - **Scale:** Large (4B–16B parameters) | |
| - **Languages:** Russian (primary), English | |
| - **License:** Apache-2.0 | |
| ### Key Features | |
| - Significantly increased model capacity compared to Orpheus-Zero | |
| - Enhanced reasoning, creativity, and instruction-following capabilities | |
| - Stronger long-context understanding | |
| - Designed as the "Emperor" variant — maximum power and dominance in generation quality | |
| ## Training & Development | |
| Orpheus-KAISER is currently in active development. The model aims to combine the clean pre-training philosophy of the Orpheus family with much greater scale and capability. | |
| **Current Status:** Pre-training / Early training phase | |
| ## Intended Use | |
| - High-quality creative writing | |
| - Complex reasoning tasks | |
| - Role-playing and world-building | |
| - Technical and creative assistance at a high level | |
| --- | |
| **"The Emperor has arrived."** | |
| **License:** Apache-2.0 | |
| **Developer:** MixlyGames (OwlNestTeam) |