jeff: a self-hosted Jev API on GLiFormer

A drop-in for TypeSafe's API powered by a 400M GLiFormer model; point the official SDK's base URL at it. Deploys on Modal or CPU via ONNX Runtime and ships a bench folder.

Notes

151 ms sequential latency vs Jev 129 ms; about $2.6 per million requests vs ~$15.6 for Jev; AG News 75.5% vs Jev 90.5%.

source-backed — Public repository, docs or live artifact. About this label

Destinations

Also linked

• Bench results ↗

Filed under

• #replica
• #self-hosted
• #github

In collections

• Local & replica models →