AWS Machine Learning Blog · Naomi Teng · 4 小时前模型评分49
Amazon Bedrock 为 Kimi K3 推出印度境内推理
Introducing India-based inference for Kimi K3 on Amazon Bedrock
导读
Amazon Bedrock 现支持 Kimi K3 在印度境内推理，满足严格数据驻留要求的本地工作负载可直接调用。Kimi K3 是 Moonshot 人工智能（AI，Artificial Intelligence） 的多模态开放权重基础模型，用户可通过 in.moonshotai.kimi-k3 推理配置文件将处理限制在孟买和海得拉巴区域。
分类置信度
中 · 待人工复核
- 存在未解决的相邻类别：ai-tooling、foundation-models，需要人工复核。
来源级别：官方一手
应用领域：医疗与生命科学、金融、政务与公共事务
事件类型：产品更新
证据摘录
- “In addition to global and US-based inference, Amazon Bedrock now supports in-country inference in India for Kimi K3: unblocking local workloads with strict data residency requirements.”
- “For less latency-sensitive cases, you can select the flex tier to enjoy a 50 percent discount on default standard pricing.”
- “For latency-critical scenarios, you can select “priority” to request higher queue priority at a higher price.”
- “Kimi K3 is billed on a per-token basis, so these code samples don’t provision any persistent AWS resources to clean up.”
- “Model invocation logging is opt-in and can optionally contain prompts and outputs.”
来源：AWS Machine Learning Blog · aws.amazon.com