When to Switch to Open Source Models: A Decision Framework
Stephane Roy
Command Palette
Search for a command to run...
Archive
10 posts published so far
How to Use EasyR1 for Reinforcement Learning on FlexAI
Stephane Roy
Running AI Workloads Across NVIDIA, AMD, and Multiple Clouds Without Refactoring
Adam King
How to Benchmark Open-Source Models Before You Commit
Stephane Roy
Fine-Tuning a 32B Legal LLM That Outperformed a Frontier Model at 4× Lower Serving Cost
Adam King
LLM Inference GPU Sizing: How to Choose the Right GPU for Your Model and Traffic
Adam King
Why AI Infrastructure Software Is Harder Than Hardware — Lessons from Building Aurora and FlexAI
Adam King
Heterogeneous AI Compute: Why Mixed NVIDIA, AMD, and TPU Clusters Are Harder Than They Look
Adam King
FlexBench: An Open-Source, Modular MLPerf Benchmark for LLM Inference
Adam King
From Supercomputers to Serverless: How FlexAI is Solving the GPU Infrastructure Challenge
Adam King