Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-09-19 · source-backed
AWS announced it September 18, describing it as the first open model at that scale, with native vision and about 2.5x the scaling efficiency of Kimi K2. It's the first open-weight model on Bedrock to support explicit prompt caching. Ships as both a US geographic profile (us.moonshotai.kimi-k3) and a global cross-Region profile (global.moonshotai.kimi-k3), the latter about 10% cheaper. AWS
Each link below shares sources, entities, or timing with this story.
Two concrete recipes for regulated customers who need inference in a single region, not merely in-geography, since cross-Region inference is the throughput-friendly default. Path one: CLAUDE_CODE_USE_MANTLE=1 plus AWS_REGION, pinning models by plain ID, supported in Ireland, S...
AWS published the arithmetic builders need to decide whether caching pays. A 10,000-token document reused across 10 questions nets roughly 75% off total input cost. Minimum checkpoint size is 1,024 tokens for Claude Sonnet 4.5 and 4.6, 4,096 for Opus. Default TTL is 5 minutes,...
The walkthrough uses utility bills queried at scale as its worked example, and argues for preprocessing before ingestion rather than a better embedding model. That matches what I hit building Rayni: the retrieval was usually fine and the extraction was wrong, and no amount of...
Bedrock invocation logs to S3 carrying model ID, token counts and IAM Identity Center user identity; an Athena view computing per-user daily spend; a Lambda on a 15-minute EventBridge schedule rewriting Customer Managed Policies via iam:CreatePolicyVersion (AWS). Denials take...
AWS shipped per-user, per-target rate limiting August 6 covering MCP targets, inference targets and HTTP passthrough across three metrics: requests (RPS/RPM), tokens (TPM, inference only), connections (CPS). Limits scope by JWT claims ($.context.jwt.sub, .role, .azp) or IAM id...
AWS made the managed AgentCore harness generally available on June 18. You define model, tools, skills, and memory with CreateHarness, then run it with InvokeHarness. It ships multi-model support (Bedrock, OpenAI, Gemini, LiteLLM), mid-session context preservation, built-in br...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.