gpt-6IMAGE
设计基准实测:只有 GPT-6 Astra 分数高于 DeepSeek V4.1 Flash
某真实设计基准上 11 个闭源模型分数更低且更贵,仅 GPT-6 Astra 得分更高,作者据此质疑闭源模型前景。
OpenDesign
@OpenDesignHQ2026.09.14收录原推发布于 2026.09.09

OpenDesign @OpenDesignHQ原推发布于 2026.09.09
2) Are closed models doomed? 11 models scored lower AND cost more than DeepSeek V4.1 Flash across our real-world design benchmark. Only GPT-6 Astra scored higher.
x.com/opendesignhq/status/2097620926556955057采集于 2026.09.14 15:48