AGI Eval
A benchmark suite for evaluating large language models on human-centric reasoning tasks, comparing model performance to average and top human performance on standardized tests like SATs, LSATs, and math competitions
As defined in Microsoft Goes FULL NUCLEAR to Develop AGI...
Where it comes up · 1
Most important claims · 1 of 1Search all claims for “AGI Eval” →
- 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.Microsoft Goes FULL NUCLEAR to Develop AGI...