模型多源确认
GPT-6 Astra 在空间推理基准测试中表现突出
GPT-6 Astra appears to show a "step change" in spatial reasoning based on early benchmarks
精选理由
OpenAI 新发布的 GPT-6 Astra 在空间推理上比 MolmoAct2 强很多，值得看看。
在 StationeryBench 基准测试中，GPT-6 Astra 使用双臂机器人完成了7个任务，而竞争对手 MolmoAct2 一个都未完成，显示出显著进步。
原文 · Decoder
GPT-6 Astra appears to show a "step change" in spatial reasoning based on early benchmarks
In a new robotics benchmark, GPT-6 Astra shows major gains in spatial understanding. On StationeryBench, the model completed 7 out of 100 tasks with dual-arm robots, while competitor MolmoAct2 couldn't finish a single one. A researcher calls it a "step change in spatial reasoning." The article GPT-6 Astra appears to show a "step change" in spatial reasoning based on early benchmarks appeared first on The Decoder .
