Amazon Web Services has added Kimi K3.1 Pro Highspeed from Moonshot AI to Amazon Bedrock. The model is a post-trained version of Kimi K3 with the same 3T-class open-weight architecture and 1-million-token context window, trained with substantially more post-training compute.
Moonshot AI reports that Kimi K3.1 Pro Highspeed generates output tokens 2–3 times faster than Kimi K3.1 Pro. According to Moonshot AI, the additional training targets long-horizon agentic work and fewer factual errors, and improves professional tasks including financial modeling, office documents, and coding for game engines, 3D scenes and computer-aided design scripts.
The model supports tool calling, structured output, configurable reasoning, image input and streaming. Kimi K3.1 Pro Highspeed uses the same Amazon Bedrock endpoint, APIs and features as Kimi K3, allowing developers already using Kimi K3 to switch by changing the model ID.
AWS offers three cross-region inference profiles: a global profile routing requests to any supported commercial AWS Region at lower per-token cost, a US profile keeping processing within US Regions, and an India profile for India Regions. Customer data is processed within the AWS data boundary, is not shared with Moonshot AI, and is not used to train the model. Zero data retention is enabled for inference requests, and AWS operators cannot view prompts and completions during inference.