Introducing India-based inference for Kimi K3 on Amazon Bedrock
Machine Learning Blog
Amazon Bedrock now supports in-country inference for Kimi K3, a multi-modal foundation model by Moonshot AI, with processing available in India's Mumbai and Hyderabad Regions to meet strict data residency requirements.
- Kimi K3 achieves frontier-level coding and general agentic intelligence capabilities
- India geographic inference profile (in.moonshotai.kimi-k3) ensures data stays within Indian Regions
- Cross-Region inference is fully managed with encrypted transit and no public internet exposure
- Supports OpenAI-compatible Chat Completions and Responses APIs for easy integration
- Example use case demonstrates OCR and structured extraction from Indic-language documents with Hinglish Q&A
- Kimi K3 handles images, PDFs, and multilingual text without separate OCR or translation services
- Prompt caching reduces costs by up to 80% and improves multi-turn chat latency by 85%
- Service tiers offer 50% discount for flex tier or priority queue access for latency-critical workloads
- CloudWatch metrics and optional model invocation logging provide usage visibility
Kimi K3 on Amazon Bedrock enables compliant AI solutions for Indian financial services and public sector workflows requiring multilingual document understanding and data residency.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
2026
2026
2026
2026
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.