rights reserved. AWS ⽣成 AI スタック APPLICATIONS TO BOOST PRODUCTIVITY MODELS AND TOOLS TO BUILD GENERATIVE AI APPS INFRASTRUCTURE TO BUILD AND TRAIN AI MODELS Amazon Q Business INS IGHT S AN D AUT OMAT ION Amazon Q Developer S OF T WARE DE VE L OP ME NT L IF E C YC L E Amazon Bedrock AMAZON MODE L S | P AR T NE R MODE L S AWS Trainium AWS Inferentia GPUs HIGH P E R F OR MANC E C OMP UT E Amazon SageMaker AI MANAGE D INF R AS T R UC T UR E
rights reserved. Amazon Confidential and Trademark. Amazon Bedrock B R O A D E S T S E L E C T I O N O F F U L L Y M A N A G E D M O D E L S F R O M L E A D I N G A I C O M P A N I E S AMAZON NOVA JAMBA CLAUDE COMMAND EMBED RERANK LLAMA LUMA RAY 2 Effective reasoning & rapid analysis for long context windows High-quality AI image generation, easily deployable at scale Advanced image & language reasoning Knowledge summarization, expert agents, & code completion High-quality video generation with natural, coherent motion & ultra-realistic details Software engineering AI for large enterprises STABLE DIFFUSION MISTRAL MIXTRAL MALIBU POINT Extended thinking & coding, including computer use skills Multimodal search & advanced retrieval powering multilingual knowledge agents Coming soon Advanced reasoning, comprehension, scientific & mathematical analysis, & code generation DEEPSEEK-R1 Frontier intelligence & industry leading price performance, high-quality image & video generation
rights reserved. Amazon Confidential and Trademark. Anthropic Claude Models on Bedrock Choose the exact combination of intelligence, speed, and cost to suit your needs 9 Claude 3.7 Sonnet Claude 3.5 Sonnet Claude 3 Haiku Claude 3 Sonnet Claude 3 Opus Use case Most intelligent & first hybrid reasoning model Most intelligent, built for high-volume use cases Fastest performance at the lowest cost Balance between intelligence, speed, and cost Second-most intelligent overall; most intelligent in Claude 3 family Context 200k 200K 200K 200K 200K Vision ✓ ✓ ✓ ✓ ✓ Cost* Input: Output: $0.003 $0.015 $0.003 $0.015 $0.00025 $0.00125 $0.003 $0.015 $0.015 $0.075 *Per 1K tokens
rights reserved. AWS ⽣成 AI スタック INFRASTRUCTURE TO BUILD AND TRAIN AI MODELS AWS Trainium AWS Inferentia GPUs HIGH P E R F OR MANC E C OMP UT E Amazon SageMaker AI MANAGE D INF R AS T R UC T UR E
rights reserved. 25 Amazon EC2 Trn2インスタンス • AWS Trainium 2 を搭載した Amazon EC2 Trn2 インスタンスが⼀般利⽤開始に • P5e/P5enと⽐較して30-40%⾼いコストパフォーマンス • ⽶国東部 (オハイオ) リージョンで Capacity Blocks for ML での提供 Instance size Trainium2 chips Chip memory Chip Memory Bandwidth vCPUs Instance Memory Storage NeuronLink EFAv3 Capacity Block Price 3Yr RI Price trn2.48xlarge 16 1.5TB 46 TB/s 192 2TB 4x 1.92TB NVMe 1 TB/s 3.2 Tb/s $35.76/hr $34.39/hr H I G H P E R F O R M A N C E training and inference of trillion+ parameter Generative AI models B E S T P R I C E - P E R F for generative AI and deep learning on AWS U P T O 4 6 T B / s of HBM Bandwidth, ideal for memory intensive token generation *2025年4⽉時点の⽶国東部 (オハイオ)の価格