A calibrated TensorRT INT8 build of RF-DETR put 0 layers in INT8 on an RTX 5090. Measured INT8, FP16, CPU quantization and distillation side by side.
AI engineer building AI agents, LLM and RAG systems, and computer vision. I write about what I test at khadim.tech.
I build AI agents, LLM and RAG systems, and computer vision pipelines, and I test new AI tools hands-on before writing about them. Top Rated Plus on Upwork. Write-ups and contact: https://khadim.tech
Markdown supported · ⌘↵ to post
No responses yet.