Liquid AI LFM2.5: On-Device MoE With 1.5B Active Params
Liquid AI released LFM2.5-8B-A1B, an on-device MoE model with 8.3B parameters but only 1.5B active at inference. 253 tok/s on M5 Max, 128K ...
Claude Code Dynamic Workflows: Run 1,000 Agents on a Single Prompt
Anthropic shipped Claude Code Dynamic Workflows on May 28. Here's the architectural shift that makes ...
Visual Studio Plan Agent: Plan Before You Code, Then Build
Visual Studio May 2026 ships a Plan agent that reads your codebase, drafts an implementation ...
Codex CLI /goal: Persistent Agent Sessions Are Now Default
Codex CLI v0.133 makes /goal persistent sessions the default. Here's what changed, how to write ...
Stable Audio 3.0: Open-Weight AI Audio You Can Actually Ship
Stable Audio 3.0 ships open weights, LoRA fine-tuning, and CPU inference. Here's the developer guide ...
DeepSeek V4: MIT Open-Source Model Matches Claude and GPT-5.5
DeepSeek V4-Pro matches Claude Opus 4.7 on SWE-bench at $1.74/M tokens under an MIT license. Benchmarks, pricing, migration guide, and data privacy tradeoffs.
OpenAI Secure MCP Tunnel: Private MCP Servers, No Firewall Changes
OpenAI shipped Secure MCP Tunnel on May 27, 2026. Connect private MCP servers to ChatGPT, ...
Cloudflare’s AI Code Review: 7 Agents, $1.19 Each
Cloudflare ran 131,246 AI code reviews in 30 days with 7 specialized agents at $1.19 ...
Google Pay & Wallet MCP Server: Connect Your AI Agent to Payment APIs
Google launched the Pay & Wallet Developer MCP server on May 27. Here is how ...
OpenAI Frontier Governance Framework: Developer Guide
OpenAI released its Frontier Governance Framework on May 28. Here's what the risk categories, Capabilities ...