Story

simon_willison ยท Jul 31, 2026 ยท news

Source brief

smevals - a small eval suite for evaluating models, prompts, and harnesses

simonwillison.netJul 31, 2026
original source linked

In brief

smevals - a small eval suite for evaluating models, prompts, and harnesses I've been working with Jesse Vincent's Prime Radiant applied AI research lab building out this evals framework to help answer questions about...

Feed lens
agentharnessevaluation

Continue reading

Read the original at simonwillison.net โ†’Open in live feedRead that dayโ€™s brief

Earlier in this thread 4 items