Open Models Almanac
Methodsinference & servinginference quantization

single mention · inference & serving

FP8 KV-cache quantization with amax-based scaling

FP8 KV-cache quantization using amax-based scaling calibrated on long-context agentic trajectories.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See inference quantization for the methods filed beside it.

Also called FP8 KV cache with amax-based calibration.

Filed under inference & serving › inference quantization