OpenRouter Fusion: They OFFICIALLY CLAIM THAT THIS MODEL beats FABLE?

In this video, I'll be talking about OpenRouter's new Fusion API, which claims to achieve Fable-level intelligence at half the price, and I'll be testing whether those claims actually hold up in real-world tasks. -- Key Takeaways: 🚀 OpenRouter has launched Fusion API, which they call the smartest compound model on the market. 🧠 Fusion works by sending prompts to multiple models in parallel and using a judge model to create the final answer. 📊 OpenRouter claims Fusion reaches Fable-level performance, but the benchmark shown is mainly focused on Deep Research tasks. ⚠️ The marketing claims seem misleading because strong Deep Research benchmark results do not prove overall model superiority. 🧪 In practical tests like simulators, SVG generation, math, and coding-style tasks, Fusion performs poorly or inconsistently. 💸 Fusion can cost more, respond slower, and sometimes produce worse results than simply using one strong model. 👍 Overall, Fusion is an interesting idea, but it does not truly surpass Fable and feels overhyped in its current state.