Open SLM Evaluations

106 Small AI Models Benchmarked — But What Can 150M Parameters Do?

Indie dev benchmarked 106 sub-150M language models on a single B200 GPU in 30 minutes. Useful starting point for local deployment, but what can 150M p