Text Generation
GGUF
Rust
English
ruvllm
agent-routing
claude-code
recursive-language-model
embeddings
llm-inference
sona
hnsw
simd
imatrix
conversational
Instructions to use Princess3/Hpp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use Princess3/Hpp with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf Princess3/Hpp:Q4_K_M # Run inference directly in the terminal: llama cli -hf Princess3/Hpp:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Princess3/Hpp:Q4_K_M # Run inference directly in the terminal: llama cli -hf Princess3/Hpp:Q4_K_M
Use pre-built binary
# Download pre-built binary from: # https://github.com/ggerganov/llama.cpp/releases # Start a local OpenAI-compatible server with a web UI: ./llama-server -hf Princess3/Hpp:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf Princess3/Hpp:Q4_K_M
Build from source code
git clone https://github.com/ggerganov/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-server llama-cli # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama-server -hf Princess3/Hpp:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf Princess3/Hpp:Q4_K_M
Use Docker
docker model run hf.co/Princess3/Hpp:Q4_K_M
- LM Studio
- Jan
- vLLM
How to use Princess3/Hpp with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Princess3/Hpp" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Princess3/Hpp", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Princess3/Hpp:Q4_K_M
- Ollama
How to use Princess3/Hpp with Ollama:
ollama run hf.co/Princess3/Hpp:Q4_K_M
- Unsloth Desktop
- Docker Model Runner
How to use Princess3/Hpp with Docker Model Runner:
docker model run hf.co/Princess3/Hpp:Q4_K_M
- Lemonade
How to use Princess3/Hpp with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Princess3/Hpp:Q4_K_M
Run and chat with the model
lemonade run user.Hpp-Q4_K_M
List all available models
lemonade list
- Atomic Chat
| { | |
| "version": "2.3", | |
| "release_date": "2026-01-20", | |
| "sota_metrics": { | |
| "total_triplets": 1078, | |
| "hard_negative_ratio": 0.484, | |
| "embedding_accuracy": 0.882, | |
| "hard_negative_accuracy": 0.812, | |
| "hybrid_routing_accuracy": 1.0, | |
| "agent_types_supported": 13 | |
| }, | |
| "training_config": { | |
| "epochs": 30, | |
| "batch_size": 32, | |
| "learning_rate": 2e-05, | |
| "loss": "triplet + infonce", | |
| "margin": 0.5, | |
| "temperature": 0.07 | |
| }, | |
| "improvements": [ | |
| "500+ Claude-generated hard negatives (up from 100)", | |
| "48% hard negative ratio (up from 18%)", | |
| "Real Candle training with gradient updates", | |
| "GRPO feedback loop with Claude-as-judge", | |
| "GGUF adapter export for llama.cpp" | |
| ] | |
| } |