// INFERENCE RUNTIME

KTransformers

Heterogeneous CPU/GPU inference for very large MoE models.
Platforms
โ€”
Backends
โ€”
Quant formats
โ€”
KTransformers website โ†—

Videos about KTransformers

Install K-Transformers to Run 600B+ AI Models Locally with Low VRAM

FAQ

What platforms does KTransformers support?โ–ถ

KTransformers runs on multiple platforms with backends various.

What quant formats does KTransformers support?โ–ถ

KTransformers supports common local formats.