Recent model released a couple of weeks ago. "Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token". Beats Kimi K2.5 and GLM 4.7 on more benchmarks than it loses to them.

Edit: there are 4 bit quants that can be run on an 128GB machine like a GB10 [1], AI Max+ 395, or mac studio.

[1] https://forums.developer.nvidia.com/t/running-step-3-5-flash...

That reverse x axis sure is confusing.