On the AGI Eval benchmark measuring human-level performance, GPT-4 scores 37.6%, Gemini 1.5 Flash scores 45%, GPT-3.5 scores 48%, and Microsoft's Grok Moe (6.6B parameters) scores 48.2%, showing that the tiny Grok Moe model is competitive with much larger models.

factualpending

Speaker

Wes Roth

Evidence Quote

“Microsoft's new model the tiny 6.6 billion parameter model did 48.2 right”

Source

Microsoft Goes FULL NUCLEAR to Develop AGI...— Wes Roth
Created: 8/10/2026, 10:52:56 PM

Your notes

Loading notes...