Devlery

Blog

Notes and analysis on AI development.

QVAC 0.12.0 brings TurboQuant to local KV cache pressure

QVAC 0.12.0 brings TurboQuant to local KV cache pressure

Tether announced QVAC SDK 0.12.0 with TurboQuant support. The useful question is how KV cache compression changes local long-context AI.

AgentCore A/B testing turns prompt edits into release experiments

AgentCore A/B testing turns prompt edits into release experiments

AWS AgentCore Optimization preview productizes agent quality loops with trace-based recommendations, batch evals, and A/B testing.

DGX Station for Windows puts trillion-parameter AI on the enterprise desk

DGX Station for Windows puts trillion-parameter AI on the enterprise desk

NVIDIA introduced DGX Station for Windows with GB300, 784GB coherent memory, and dual boot for enterprise AI agent workstations.

Microsoft expands Copilot ISO 42001 scope to Studio agents

Microsoft expands Copilot ISO 42001 scope to Studio agents

Microsoft is expanding Copilot ISO 42001 coverage to Copilot Studio, GitHub Copilot, Dragon Copilot, and Copilot Health.

AWS Nova Act Service Card defines the limits of browser agents

AWS Nova Act Service Card defines the limits of browser agents

AWS documented Nova Act limits for browser agents, including 100 sequential steps, 30-minute sessions, prompt injection boundaries, and IAM resources.

QVAC AI SDK Provider lets local models plug into AI SDK apps

QVAC AI SDK Provider lets local models plug into AI SDK apps

Tether QVAC published a Vercel AI SDK provider for local OpenAI-compatible servers. Here is what it changes for agents, TypeScript apps, and local AI routing.

Cohere Command A+ ships as an open-weight MoE for two H100s

Cohere Command A+ ships as an open-weight MoE for two H100s

Cohere Command A+ combines Apache 2.0 open weights, a 218B MoE design, 25B active parameters, 128K context, and enterprise deployment options.

SageMaker Adds OpenAI API Support for AWS-Hosted Models

SageMaker Adds OpenAI API Support for AWS-Hosted Models

AWS SageMaker now supports /openai/v1 endpoints, lowering the migration cost for OpenAI SDK, LangChain, Strands Agents, and AI gateways.

MiniMax M3 brings 1M context to open-weight coding models

MiniMax M3 brings 1M context to open-weight coding models

MiniMax M3 combines 1M context, multimodality, and coding-agent benchmarks, but its weights and technical report are still pending verification.

Mistral Search Toolkit separates RAG failures from model failures

Mistral Search Toolkit separates RAG failures from model failures

Mistral Search Toolkit public preview treats RAG and agent-search failures as retrieval, pipeline, and evaluation problems.

RTX Spark Debuts as a Local AI PC for 120B LLMs

RTX Spark Debuts as a Local AI PC for 120B LLMs

NVIDIA and Microsoft introduced RTX Spark, a Windows PC category for local agents with 120B LLMs, 128GB unified memory, and OpenShell.

Codex Tax AI handled 7,000 returns, and the improvement loop starts with evals

Codex Tax AI handled 7,000 returns, and the improvement loop starts with evals

OpenAI and Thrive showed how Tax AI links production traces, practitioner corrections, evals, and Codex tasks.