SRStephane Royinflexai.hashnode.dev·May 11 · 15 min readWhen to Switch to Open Source Models: A Decision FrameworkCompare cost, control, privacy, and workload fit before you move from closed-source APIs to open source LLMs. Most teams paying for OpenAI, Gemini, or Claude APIs eventually hit the same question: are00
SRStephane Royinflexai.hashnode.dev·May 8 · 15 min readHow to Use EasyR1 for Reinforcement Learning on FlexAIEasyR1 is a reinforcement learning fine-tuning framework that supports GRPO, DAPO, and REINFORCE for reasoning-focused post-training. Use it when SFT starts plateauing on tasks like math, code, or log00
SRStephane Royinflexai.hashnode.dev·Apr 27 · 5 min readHow to Benchmark Open-Source Models Before You CommitYou're choosing between Llama 4 Scout 17B, GPT-OSS 120B, and DeepSeek V3.2. The paper numbers look fine across all three. You pick the one that feels right and ship it. Three weeks later it fails on t00
