模型精选
DeepSeek V4.1-Flash 模型 KV 缓存优化
💾 Smaller KV cache. Bigger savings. Compared with the previous generation, V4.1-Flash’s KV cache n...
精选理由
DeepSeek 新模型 KV 缓存更小，能省下不少钱，适合做代理。
DeepSeek 新发布的 V4.1-Flash 模型，其 KV 缓存需求仅为上一代的 1/4 HBM 和 1/8 SSD 存储，能显著降低代理成本。
原文 · 深度求索 DeepSeek
💾 Smaller KV cache. Bigger savings. Compared with the previous generation, V4.1-Flash’s KV cache n...
💾 Smaller KV cache. Bigger savings. Compared with the previous generation, V4.1-Flash’s KV cache needs just: 🔹 1/4 the HBM 🔹 1/8 the SSD storage Cache-hit charges often account for a large share of agent costs. Compressing the cache cuts those costs significantly. 3/6 💬 11 🔄 61 ❤️ 1192 👀 96316 📊 127 ⚡
