Google has released Agent Substrate 1.0, an open-source platform designed to handle AI agent sessions on Kubernetes. By separating stateful sessions from compute resources, it allows idle agents to be 'parked' in object storage, freeing up worker nodes.
Amazon has released a reference architecture that integrates the NVIDIA NeMo Switchyard library with LiteLLM to dynamically route coding agent requests between different models on Bedrock. By using stage or random routing, you can shift traffic between high-capability and high-efficiency models to balance performance and cost.
Amazon has published a workflow for using Sequence-Level Knowledge Distillation (SeqKD) to fine-tune smaller models on trajectories generated by frontier models. By using a 'teacher' model to generate successful tool-use trajectories and filtering them with a programmatic scorer, teams can train compact 'student' models that perform complex reasoning at a fraction of the cost.
AWS published a walkthrough for the Adaptive Agentic Intelligence Framework (AAIF), built on Strands Agents, which routes simple requests to small local models (via Ollama) and only sends complex requests to cloud models through Amazon Bedrock. A lightweight 1B-parameter model (llama3.2:1b) acts as an orchestrator deciding where each request should go, with an example IoT monitoring agent that uses local telemetry retrieval, local vision analysis, and cloud-based expert insights. AWS states that in the hybrid setup, roughly 80 percent of traffic can be handled locally without hitting a paid API, and the architecture can run fully on edge hardware (e.g. Raspberry Pi with 8GB memory), fully in the cloud, or in a hybrid mode.
OpenAI's Responses API now supports a 12-hour prompt cache option, keeping shared instructions, tools and context cached for longer so repeated requests hours apart can still benefit from cheaper cached-input pricing. Writing to this longer cache costs 2.5x the standard input token price. This is an optional setting alongside the existing shorter default cache.
Cohere built a controlled testbed to study how teams of coding agents should be organized, testing where work is delegated, how leads wait on workers, which models play which role, team size, and communication patterns. On ProgramBench, well-organized teams outperformed single agents by up to ~10 percentage points but at 3–7x higher cost. The key finding: delegating only exploration and planning to workers matched the performance of delegating all stages (implementation and verification too) at half the cost, and a strong lead paired with 16 cheap planning workers reached 61.8% reward versus 63.2% for an all-strong-model team, cutting cost from $510 to $155.
Apple ML researchers published a NeurIPS paper introducing SELFCOMPACT, a scaffold that lets language model agents decide for themselves when and how to compact their own accumulated context, rather than relying on fixed-interval summarization triggered by a token count. It combines a compaction tool the model can call with a lightweight rubric telling it when to fire (a sub-task is resolved) and when to hold off (mid-derivation or stuck). Tested on six benchmarks and seven models, it matched or beat fixed-interval summarization while using far fewer tokens.
Google Cloud announced the Gemini agent, a unified agent that combines chat, knowledge work, content creation, and coding into one interface accessible across devices and channels. It integrates directly inside Gmail, Docs, Sheets, Slides, Chat, and Calendar, connects to enterprise tools and databases via MCP and a tools registry, and orchestrates across multiple models — including Google's Gemini family and Anthropic's Claude — to balance quality and cost. Google also detailed security/governance features, cost controls like Smart Routing and spend caps, and shared customer results from companies like BNP Paribas, Bradesco, and Orange Spain.