3 ms·Gemini 2.5: Updates to our family of thinking models4 points by meetpateltech 1y agodrewbitt 1y agoI would expect 2.5 Flash-Lite to take over the translation scene like 1.5 and 2.0 did at that price and latency (280-320 token/s right now)