大宇|Nov 19, 2026 10:49
Google's new model, Neon, is not a flagship model, but a mid-range mainstay.
According to Google's own testing, when reading a million words at once and then answering the relationships within them, Neon scored 95.2, Opus 5.5 scored 66.8, and GPT-6.2 Sol scored 80.1.
Google positions Carbon for the most challenging long-duration tasks, hoping Neon will become a primary programming tool, competing with Opus.
Neon has brought Google's mainstay model back into the top tier.
Share To
Timeline
HotFlash
APP
X
Telegram
CopyLink