---
title: "Stephane Roy (@steph1neproflex) | Hashnode"
description: "Stephane Roy (@steph1neproflex) writes on Hashnode. Read their latest articles and follow their work."
url: "https://hashnode.com/@steph1neproflex"
---

@steph1neproflex

# Stephane Roy

@steph1neproflex · Barcelona, Spain · Joined April 2026

### [When to Switch to Open Source Models: A Decision Framework](https://flexai.hashnode.dev/when-to-switch-to-open-source-models?utm_source=hashnode&utm_medium=feed)

hen should you move from closed source models to open source models? Compare cost, control, privacy, and workload fit first.

15 min read · [flexai.hashnode.dev](https://flexai.hashnode.dev/when-to-switch-to-open-source-models?utm_source=hashnode&utm_medium=feed)

[May 11](https://hashnode.com/posts/when-to-switch-to-open-source-models/6a0193a11c751823747b5eb9)

### [How to Use EasyR1 for Reinforcement Learning on FlexAI](https://flexai.hashnode.dev/how-to-use-easyr1-for-reinforcement-learning-on-flexai?utm_source=hashnode&utm_medium=feed)

EasyR1 for reinforcement learning: compare GRPO vs DAPO, understand the setup and cost, and run your first RL fine-tuning workflow on FlexAI.

15 min read · [flexai.hashnode.dev](https://flexai.hashnode.dev/how-to-use-easyr1-for-reinforcement-learning-on-flexai?utm_source=hashnode&utm_medium=feed)

[May 8](https://hashnode.com/posts/how-to-use-easyr1-for-reinforcement-learning-on-flexai/69fe0bf1f239332df4e0fdf8)

### [How to Benchmark Open-Source Models Before You Commit](https://flexai.hashnode.dev/how-to-evaluate-open-source-models?utm_source=hashnode&utm_medium=feed)

A guide to lm-evaluation-harness — which benchmark tasks matter for your use case, how to run them, and how to interpret results before you commit.

5 min read · [flexai.hashnode.dev](https://flexai.hashnode.dev/how-to-evaluate-open-source-models?utm_source=hashnode&utm_medium=feed)

[Apr 27](https://hashnode.com/posts/how-to-evaluate-open-source-models/69ef6f5d330a1ad7f7ea8433)
