> ## Documentation Index > Fetch the complete documentation index at: https://fal.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Optimizing Cold Starts > Understanding cold starts and how to reduce them. ## Cold Starts vs Warm Starts A **cold start** occurs when a new runner needs to be created from scratch. The runner goes through `PENDING` → `SETUP` → `IDLE` (or `PENDING` → `DOCKER_PULL` → `SETUP` → `IDLE` if the Docker image isn't [cached](/documentation/deployment/caching)) before it can serve requests. A **warm start** occurs when an existing IDLE runner is reused to handle a new request: `IDLE` → `RUNNING`. ### What Triggers Cold Starts * No warm runners available (all busy or expired) * Traffic spike exceeds warm runner capacity * First deployment * Runners expired during low traffic periods ### Factors Affecting Cold Start Duration * **Image size**: Larger Docker images take longer to pull * **Model size**: Larger models take longer to download and load * **Setup complexity**: Complex initialization in `setup()` adds time * **Cache state**: First runs are slower, subsequent runs benefit from caching * **Hardware availability**: GPU availability varies by region and time ## How to Reduce Cold Starts Each of these strategies targets a different phase of the cold start: Keep warm runners available with keep\_alive, min\_concurrency, and buffers Reduce image size for faster pulls with multi-stage builds and smaller base images Cache compiled kernels to speed up setup() across runners Download models to /data for automatic caching between runners