hybrid-inference
Krasis is a Hybrid LLM runtime which focuses on efficient running of larger models on consumer grade VRAM limited hardware
Official Python SDK for the TQNN Fault-Tolerant Inference Platform.
