emfhal/ gpt-oss

632 7 months ago

CPU-optimized gpt-oss:20b for Kubernetes and VM environments. Runs without GPU, tuned for predictable CPU inference, balanced quality, and reduced memory usage.

tools thinking

1 model

gpt-oss:20b-cpu-optimized
5b8810a7433d • 14GB • 128K context window •
Text input • 7 months ago