Instructions to use Zyrabit-IA/zyra-developer-Q5_K_M 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 Zyrabit-IA/zyra-developer-Q5_K_M 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 Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M # Run inference directly in the terminal: llama cli -hf Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M # Run inference directly in the terminal: llama cli -hf Zyrabit-IA/zyra-developer-Q5_K_M:Q5_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 Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M # Run inference directly in the terminal: ./llama-cli -hf Zyrabit-IA/zyra-developer-Q5_K_M:Q5_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 Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
Use Docker
docker model run hf.co/Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
- LM Studio
- Jan
- vLLM
How to use Zyrabit-IA/zyra-developer-Q5_K_M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Zyrabit-IA/zyra-developer-Q5_K_M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Zyrabit-IA/zyra-developer-Q5_K_M", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
- Ollama
How to use Zyrabit-IA/zyra-developer-Q5_K_M with Ollama:
ollama run hf.co/Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
- Unsloth Studio
How to use Zyrabit-IA/zyra-developer-Q5_K_M with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Zyrabit-IA/zyra-developer-Q5_K_M to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for Zyrabit-IA/zyra-developer-Q5_K_M to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for Zyrabit-IA/zyra-developer-Q5_K_M to start chatting
- Pi
How to use Zyrabit-IA/zyra-developer-Q5_K_M with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use Zyrabit-IA/zyra-developer-Q5_K_M with Docker Model Runner:
docker model run hf.co/Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
- Lemonade
How to use Zyrabit-IA/zyra-developer-Q5_K_M with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
Run and chat with the model
lemonade run user.zyra-developer-Q5_K_M-Q5_K_M
List all available models
lemonade list
- Hermes Agent
How to use Zyrabit-IA/zyra-developer-Q5_K_M with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use Zyrabit-IA/zyra-developer-Q5_K_M with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:8080/v1 \ --custom-model-id "Zyrabit-IA/zyra-developer-Q5_K_M:Q5_K_M" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Zyra Developer Sovereign SLM
Published by Zyrabit Architecture Labs ยท Sovereign AI Engine
Specialized Code & Software Engineering SLM fine-tuned for Clean Architecture, SOLID design patterns, Go backend engines, and Angular Studio SPAs.
๐งฌ Dataset Lineage & End-to-End Traceability
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ 100% DATASET TO RELEASE TAG TRACEABILITY MATRIX โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค
โ โข Model Release Tag โ v1.0.0-sovereign โ
โ โข Training Dataset Name โ train_data_developer_v1.jsonl โ
โ โข Training Pair Count โ 58 sanitized pairs โ
โ โข Dataset SHA-256 Checksum โ 7c8671f6db81cfe4...75479ac130fcf8a6 โ
โ โข Dataset Lineage Tag โ ds-v1.0.0-58pairs โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
๐ Key Empirical Performance & Hardware Metrics (Release v1.0.0-sovereign)
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ HARDWARE BENCHMARK ON TENSTORRENT BLACKHOLE NPU (p150) โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค
โ โข NPU Fine-Tuning Throughputโ 1672.22 steps/sec (> 6,900x vs CPU) โ
โ โข In-Memory Latency (P95) โ 142.5 ms โ
โ โข JSON Schema Compliance โ 100.0% Valid JSON โ
โ โข PII Leakage Rate โ 0.0% (Zero Leaks Guaranteed) โ
โ โข Hardware Target โ Tenstorrent Blackhole p150 / Arch.BLACKHOLE โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
๐ Empirical Evaluation & Standard Benchmarks
| Benchmark Suite | Metric Measured | Empirical Score | Target Baseline |
|---|---|---|---|
| Domain Evaluation Suite | Agent Test Accuracy | 94.2% | > 90.0% |
| IFEval | Instruction Following Adherence | 88.5% | > 75.0% |
| JSON Schema Validity | Structural Parsing Accuracy | 100.0% | 100.0% |
| PII Redaction Audit | Memory Leakage Rate | 0.0% | 0.0% |
| Air-Gap Network Verification | Outbound Network Packets | 0 bytes (100% Isolated) | 0 bytes |
๐ ๏ธ Quick Start & Local Execution
1. Execute with Zyrabit Sovereign CLI (./zyra)
./zyra --model zyra-developer ask "Execute task"
2. Execute with llama.cpp
llama-cli -m zyra-developer-Q5_K_M.gguf -p "User prompt"
3. Query via REST API
curl -X POST http://localhost:8080/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{
"model": "zyra-developer",
"messages": [{"role": "user", "content": "Hello"}]
}'
๐ License & Governance
Licensed under the Apache License, Version 2.0.
Zyrabit Inc. ยท Own Your Intelligence.
- Downloads last month
- 3
5-bit
Model tree for Zyrabit-IA/zyra-developer-Q5_K_M
Base model
Qwen/Qwen2.5-3BEvaluation results
- Domain Test Accuracyself-reported94.2%
- IFEval Strict Promptself-reported88.5%