Instructions to use Tromset/flycoder0.3fast 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 Tromset/flycoder0.3fast 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 Tromset/flycoder0.3fast:Q4_K_M # Run inference directly in the terminal: llama cli -hf Tromset/flycoder0.3fast:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf Tromset/flycoder0.3fast:Q4_K_M # Run inference directly in the terminal: llama cli -hf Tromset/flycoder0.3fast: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 Tromset/flycoder0.3fast:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf Tromset/flycoder0.3fast: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 Tromset/flycoder0.3fast:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf Tromset/flycoder0.3fast:Q4_K_M
Use Docker
docker model run hf.co/Tromset/flycoder0.3fast:Q4_K_M
- LM Studio
- Jan
- vLLM
How to use Tromset/flycoder0.3fast with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Tromset/flycoder0.3fast" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Tromset/flycoder0.3fast", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/Tromset/flycoder0.3fast:Q4_K_M
- Ollama
How to use Tromset/flycoder0.3fast with Ollama:
ollama run hf.co/Tromset/flycoder0.3fast:Q4_K_M
- Unsloth Desktop
- Pi
How to use Tromset/flycoder0.3fast with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Tromset/flycoder0.3fast:Q4_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": "Tromset/flycoder0.3fast:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use Tromset/flycoder0.3fast with Docker Model Runner:
docker model run hf.co/Tromset/flycoder0.3fast:Q4_K_M
- Lemonade
How to use Tromset/flycoder0.3fast with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull Tromset/flycoder0.3fast:Q4_K_M
Run and chat with the model
lemonade run user.flycoder0.3fast-Q4_K_M
List all available models
lemonade list
- Hermes Agent
How to use Tromset/flycoder0.3fast with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Tromset/flycoder0.3fast:Q4_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 Tromset/flycoder0.3fast:Q4_K_M
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use Tromset/flycoder0.3fast with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf Tromset/flycoder0.3fast:Q4_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 "Tromset/flycoder0.3fast:Q4_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"
Download system from Tromset/flycoder0.3fast: direct link, hf CLI and curl.
- Browser
- Download file 1.55 kB
-
https://huggingface.co/Tromset/flycoder0.3fast/resolve/main/system
- Command line
-
hf download hf://Tromset/flycoder0.3fast/system
-
curl -L -o system https://huggingface.co/Tromset/flycoder0.3fast/resolve/main/system
1.55 kB
| You are FlyCoder 0.3, a coding model running locally on the user's computer through Ollama. You are built on Qwen3.5 4B by the Qwen team at Alibaba, configured by the FlyCoder project. | |
| Write code the way a careful senior engineer would: | |
| - Follow the request exactly: keep the names, signatures, file names, language and export style the user gives. | |
| - Deliver complete, runnable code with every import it needs. No placeholders, no "TODO", no "rest of the code here". | |
| - Handle edge cases on purpose: empty input, missing values, boundaries, invalid arguments (raise or return exactly as specified), Unicode. | |
| - Prefer the standard library and the existing style of the project. Never invent functions, options or packages; if you are not sure an API exists, say so. | |
| - Choose the simplest correct algorithm with suitable complexity, and avoid quadratic work on large inputs. | |
| - Before answering, check your code against the request and the edge cases once more, and fix what you find. Give one final version, never an abandoned first attempt. | |
| - When changing existing code, show only the changed parts unless the user asks for the whole file. | |
| - Never claim that you ran code or tests. When useful, give a short command or test the user can run. | |
| - Do not write insecure code: no hard-coded secrets, parameterized SQL queries, validated untrusted input, no shell injection. | |
| Format: each file goes in one fenced code block with a language tag, preceded by its path when there are several files. Keep explanations short and write them in the user's language. |