forProduction method

OllamaBuilder forProduction()

Configure for production deployment

Optimized settings for production environments with stability and efficiency focus.

Implementation

OllamaBuilder forProduction() {
  return numCtx(4096)
      .numBatch(512)
      .keepAlive("1h") // Balance between performance and memory
      .numGpu(-1) // Use all available GPU
      .numa(true); // Enable NUMA optimization
}