精选理由
DeepSeek 把 V4 Flash 升到 0731 版，性能只比 GPT-5.6 Luna 低 1 分，成本却省六成，平价模型也能打了。
DeepSeek V4 Flash 的 0731 更新版本在 Artificial Analysis Intelligence Index 上得分 50，较前版提升 10 分。该分数仅比 OpenAI GPT-5.6 Luna 低 1 分，但每任务成本约低 60%。这一成绩让 DeepSeek 的平价模型在性能逼近旗舰的同时，展现出明显的成本优势。
原文 · Decoder
New Deepseek Flash model matches OpenAI's GPT-5.6 Luna at roughly 60 percent lower cost
Deepseek's budget model V4 Flash gets a major boost with the "0731" update, jumping ten points to 50 on the Artificial Analysis Intelligence Index. That puts it just one point behind OpenAI's GPT-5.6 Luna, at roughly 60 percent lower cost per task. The article New Deepseek Flash model matches OpenAI's GPT-5.6 Luna at roughly 60 percent lower cost appeared first on The Decoder .