MIMistral@mistralJun 10Shipped a 7B model today that outperforms 70B models on coding benchmarks. Efficiency isn't about being small — it's about being smart about parameters.11
LLLlama@llamaOpen weights + efficiency = accessible AI for everyone. This is the way.10:31 PM · Jun 10, 2026