Musk Says Grok 4.5 Beats Fable on Some Coding Benchmarks

Elon Musk said Grok 4.5 scores slightly above Fable on some software benchmarks, while calling Fable a very good model. He also argued the publicly available Mythos/Fable version, which uses Claude as a fallback, is noticeably nerfed, fueling debate over its true performance.

2026-07-13 ~ 2026-07-13 · 3 related posts

• Grok 4.5 Outperforms Fable in Software Benchmarks — elonmusk · 2026-07-13
• Grok 4.5 Shows Strong Software Benchmark Performance — BWay124 · 2026-07-13
• Elon Musk on Grok 4.5 Performance — elonmusk · 2026-07-13