nebuly-ai/optimate
OptiMate is a collection of libraries for optimizing AI model inference performance on GPUs and CPUs.

Velocity · 7d
+5.3
★ / day
Trend
→steady
star history
OptiMate is an open-source toolkit for AI model optimization developed by Nebuly AI. It provides Speedster for inference cost reduction using hardware-aware optimization techniques, Nos for dynamic GPU cluster partitioning in Kubernetes, and ChatLLaMA for fine-tuning optimization with RLHF alignment. The repository is currently in legacy status and not actively maintained.