At 740M parameters, it’s competitive across benchmarks – even outperforming some specialist models more than twice its size.
Developers can use it to add multimodal search to their apps – like finding moments in a video using a voice memo – or pair it with Gemma 4 for private,



原文链接:https://x.com/GoogleDeepMind/status/2107502289652982118
中文翻译(AI 生成,仅供参考)
在 740M 参数下,它在各项基准测试中具有竞争力——甚至超越了一些规模是其两倍以上的专用模型。
开发者可以利用它为应用添加多模态搜索——比如通过语音备忘录在视频中查找片段——或将其与 Gemma 4 结合用于私有化,