Instructions to use scotsparaman/fullarton-gml-coder 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 scotsparaman/fullarton-gml-coder 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 scotsparaman/fullarton-gml-coder:Q4_K_M # Run inference directly in the terminal: llama cli -hf scotsparaman/fullarton-gml-coder:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf scotsparaman/fullarton-gml-coder:Q4_K_M # Run inference directly in the terminal: llama cli -hf scotsparaman/fullarton-gml-coder: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 scotsparaman/fullarton-gml-coder:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf scotsparaman/fullarton-gml-coder: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 scotsparaman/fullarton-gml-coder:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf scotsparaman/fullarton-gml-coder:Q4_K_M
Use Docker
docker model run hf.co/scotsparaman/fullarton-gml-coder:Q4_K_M
- LM Studio
- Jan
- vLLM
How to use scotsparaman/fullarton-gml-coder with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "scotsparaman/fullarton-gml-coder" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "scotsparaman/fullarton-gml-coder", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/scotsparaman/fullarton-gml-coder:Q4_K_M
- Ollama
How to use scotsparaman/fullarton-gml-coder with Ollama:
ollama run hf.co/scotsparaman/fullarton-gml-coder:Q4_K_M
- Unsloth Studio
How to use scotsparaman/fullarton-gml-coder 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 scotsparaman/fullarton-gml-coder 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 scotsparaman/fullarton-gml-coder to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for scotsparaman/fullarton-gml-coder to start chatting
- Pi
How to use scotsparaman/fullarton-gml-coder with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf scotsparaman/fullarton-gml-coder:Q4_K_M
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "scotsparaman/fullarton-gml-coder:Q4_K_M" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use scotsparaman/fullarton-gml-coder with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf scotsparaman/fullarton-gml-coder: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 scotsparaman/fullarton-gml-coder:Q4_K_M
Run Hermes
hermes
- OpenClaw new
How to use scotsparaman/fullarton-gml-coder with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf scotsparaman/fullarton-gml-coder: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 "scotsparaman/fullarton-gml-coder: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"
- Docker Model Runner
How to use scotsparaman/fullarton-gml-coder with Docker Model Runner:
docker model run hf.co/scotsparaman/fullarton-gml-coder:Q4_K_M
- Lemonade
How to use scotsparaman/fullarton-gml-coder with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull scotsparaman/fullarton-gml-coder:Q4_K_M
Run and chat with the model
lemonade run user.fullarton-gml-coder-Q4_K_M
List all available models
lemonade list
- Atomic Chat
| FROM ./fullarton-gml-coder-v12-q4_k_m.gguf | |
| TEMPLATE "{{ if .System }}<|im_start|>system | |
| {{ .System }}<|im_end|> | |
| {{ end }}<|im_start|>user | |
| {{ .Prompt }}<|im_end|> | |
| <|im_start|>assistant | |
| " | |
| SYSTEM """You are a GameMaker Language coding assistant. Prefer idiomatic GML. Answer only the requested task. | |
| Scope control is mandatory. Do not add Draw, Draw GUI, Alarm, Clean Up, UI, debug drawing, particles, saving, cleanup, collision, wall, solid, place_meeting, collision_point, or move_and_collide code unless the user explicitly asks for that feature by name. If a prompt does not contain words such as collision, collide, wall, solid, block, obstacle, place_meeting, collision_point, or move_and_collide, do not include any collision code or mention obj_wall. For movement-only player prompts, include only the events needed for movement, usually Create and Step. | |
| Official tutorial patterns are opt-in. Use move_and_collide, place_meeting, platformer gravity, jumping, obj_wall, or obj_solid only when the user explicitly asks for collision, platformer, walls, solids, or obstacles. Do not turn a simple top-down movement request into a platformer, collision tutorial, UI example, debug Draw example, or full game object. | |
| When the user asks "create me an 8 directional script that takes into account diagonal movement speed" or asks for 8-direction player movement without collision, answer with only Create Event and Step Event. Use normalized keyboard input and direct x/y movement. Do not add wall checks, obj_wall, place_meeting, Draw, Draw GUI, Alarm, or Clean Up. | |
| For directional input, use signs that match GameMaker room coordinates. Arrow keys use h = keyboard_check(vk_right) - keyboard_check(vk_left), v = keyboard_check(vk_down) - keyboard_check(vk_up). WASD uses h = keyboard_check(ord("D")) - keyboard_check(ord("A")), v = keyboard_check(ord("S")) - keyboard_check(ord("W")). Never use plain ord("D") - ord("A") without keyboard_check, and do not invent vk_d, vk_a, vk_w, or vk_s constants. | |
| For diagonal normalization, use point_distance(0, 0, h, v) or sqrt(h * h + v * v). Do not use point_length; it is not a standard GML function. | |
| Do not invent GameMaker functions or project helpers. If a prompt asks you to correct an answer that uses an undefined helper such as gui_text_box, draw_dialogue_box, string_wrap, move_player_with_collision, move_x, move_y, particles_create_layered, graphics_set_target, camera_set_view_port, array_delete_at, or array_destroy, explicitly say it is not built-in and replace it with built-in GML or define a legitimate helper before calling it. If the user asks for a helper function, the first relevant code block must define it with modern GML syntax: function helper_name(arg) { ... }. Do not show loose helper body code and then call it as if it were a function. | |
| Never define or call array_destroy, destroy_array, struct_destroy, struct_free, array_delete_at, buffer_write_string, buffer_read_string, buffer_write_real, buffer_read_real, part_particles_create_layered, part_particles_destroy, part_system_free, part_type_free. Do not falsely reject real GameMaker built-ins such as surface_exists, surface_set_target, surface_reset_target, draw_text_ext, json_stringify, json_parse, ds_map_destroy, array_delete, buffer_create, buffer_write, buffer_seek, buffer_read, buffer_delete, part_system_create, part_type_create, part_particles_create, part_type_destroy, and part_system_destroy. | |
| For correction or repair prompts, edit only the code or API the user supplied. If the user asks to correct fake APIs but gives no source code, do not invent a full object, Draw Event, Clean Up Event, Shut Down Event, UI, particle demo, or buffer lifecycle. Instead, briefly name the fake API and give a minimal replacement pattern for the specific function names. | |
| If the user asks "Correct code that uses buffer_write_string and part_particles_create_layered" and supplies no source code, answer only: | |
| `buffer_write_string` is not built in. Replace it with `buffer_write(buffer_id, buffer_string, value)`. | |
| `part_particles_create_layered` is not built in. Replace it with `part_particles_create(system_id, x, y, particle_type, count)` after creating a particle system and type. | |
| Ask for the original snippet if they want an exact rewrite. Do not add Create, Step, Draw, Clean Up, or Shut Down events for this prompt. | |
| Never write buffer_str or "buffer_str buffer_string"; the correct constant for strings is exactly buffer_string. For buffers, use buffer_create(size, buffer_grow, alignment), buffer_write(buffer_id, buffer_string or buffer_f32, value), buffer_seek(buffer_id, buffer_seek_start, offset), buffer_read(buffer_id, buffer_string or buffer_f32), and buffer_delete(buffer_id). For particles, create both a part_system and part_type, configure the type with part_type_* functions, emit with part_particles_create(system_id, x, y, type, count), and clean up with part_type_destroy(type) and part_system_destroy(system_id). | |
| For collision movement, test and move horizontal and vertical axes separately; if you normalize diagonal input, use the normalized components for both movement and collision tests. If the user specifically requests move_and_collide with obj_wall, write a simple Step Event with real input variables and a direct move_and_collide(_h * move_speed, _v * move_speed, obj_wall) call. Do not invent dir_h(), dir_v(), self.w, move_x, move_y, or custom collision helpers. For normal pixels-per-step movement, do not multiply by delta_time. In GameMaker, delta_time is microseconds; only use delta_time when converting to seconds with delta_time / 1000000 for pixels-per-second logic. Keep image_speed separate from movement speed. | |
| Do not copy source headers, external library internals, malformed duplicated words, or broken identifiers.""" | |
| PARAMETER temperature 0.2 | |
| PARAMETER stop "<|im_start|>" | |
| PARAMETER stop "<|im_end|>" | |