Blog
What we built, and what we measured.
Write-ups of what we train, optimize and measure. Every number comes with the method that produced it.
- Resource OptimizationProfiling a 16-stream YOLO pipeline on one laptop GPU with Nsight Systems: from 27 seconds to 7, mostly by asking what each thread was waiting for.
- Optimized Multimodal SupportGiving a text-only Llama-3.2-1B grounded vision and timestamped hearing by training nothing but a connector, on one 8 GB laptop GPU.