快讯
Liquid AI 发布 Open d1 系列两个开源权重多模态决策模型
Liquid AI 宣布发布 Open d1：其 d1 决策模型家族的两个开源权重多模态模型。团队表示，d1-3B 支持文本+视觉，d1-omni-600M 支持文本+图像或文本+音频；转发者称在 NVIDIA GeForce RTX 4090 上 8 毫秒即可完成决策，官方称可从 NVIDIA DGX 数据中心到 RTX 工作站再到 Jetson 边缘设备实现实时决策。
查看原文
RT Sina Semnani Our first two open-weight decision models are out and they are multimodal! Decisions in just 8 ms on an NVIDIA GeForce RTX 4090. Liquid AI: Today we release Open d1: two open-weight multimodal models in our d1 decision model family. > d1-3B: text + vision > d1-omni-600M: text + image or text + audio > Real-time decision making anywhere, from data centers such as @nvidia DGX to RTX workstations to Jetson at the edge.
前后快讯
上一篇研究：模型不变时，agent harness 可带来最高 3 倍成本差异
下一篇Scale Labs 推出 Humanity’s Sixth Sense 视觉推理基准：人类得分 93.1%，最强模型仅 53.6%