以下部分引用 AI 总结，现阶段 AI 仍然有幻觉。请以内容的原文为准。
数据新鲜度群聊 07-14 ✓卖方 07-14 ✓Wrap 断档15天 ✕News 07-15 ✓X 断档12天 ✕页面生成 07-15
Gemini Flash model
4 atoms · 跨 2 天 · 首见 2026-05-19 · 最近 2026-05-20
三色: 🟦 fact 4 · 🟥 take 0 · stance ▲0/▼0/◆0
来源: X 4
时态: fresh:4
标签: 好数字:3 · 好观点:1
事实 (fact) (4)
- 2026-05-19 · inference speed on TPU 8i
Google just showed a demo, Gemini Flash model running between 600-1400 tokens per second on TPU 8i
tags: 好数字 · → daily
value: qty=600-1400 tokens per second · date=2026-05-19
- 2026-05-19 · peak inference speed on TPU 8i
It peaked out around 1480 tok/s
tags: 好数字 · → daily
value: qty=1480 tokens per second
- 2026-05-19 · average inference speed on TPU 8i
with average around 800 tok/s
tags: 好数字 · → daily
value: qty=800 tokens per second
- 2026-05-20 · speed context
This is per user speed
tags: 好观点 · → daily
value: date=2026-05-20
时间轴 (近 20)
- 2026-05-20 ·
fact · speed context · → daily
- 2026-05-19 ·
fact · inference speed on TPU 8i · → daily
- 2026-05-19 ·
fact · peak inference speed on TPU 8i · → daily
- 2026-05-19 ·
fact · average inference speed on TPU 8i · → daily
← 实体目录 · 系统日志