Ideas for building
better software.

AI Gateway: GPT-5.6 pricing and speed updates

On , and are now cheaper and is faster.AI GatewayGPT-5.6 LunaGPT-5.6 TerraGPT-5.6 Sol AI Gateway adds no markup on token pricing, so these changes reach you at the upstream rate. The changes apply to both short and long context pricing. GPT-5.6 Sol keeps the same price; its fast mode now runs 2.5x faster, up from 1.5x. Model IDs are unchanged, so existing requests get the new rates and speed with no code change. See full pricing detail on .AI Gateway Read more Model Change Input: Short co

Deployments are now up to 7 seconds faster

Over the past few months, we've made deployments up to 7 seconds faster end to end. We removed up to 5 seconds of fixed platform overhead from every build, and deploys from the latest Vercel CLI save up to 2 seconds more. The improvements are most noticeable on smaller builds, where orchestration represents a larger share of the total build time. The largest platform gains: These apply automatically to builds triggered through Git, the dashboard, or Vercel CLI. Deployments from the latest CLI fi

Latest MCP spec now supported in mcp-handler

is now available with support for the and [email protected] Model Context Protocol specificationMCP TypeScript SDK v2 We originally built to make it easier to spin up MCP servers on popular web frameworks including Next, Nuxt, Svelte, and more.mcp-handler With the new 2.0 release, the handler now supports: Existing stateless Streamable HTTP clients can continue connecting to the same endpoint while servers adopt the latest protocol./mcp Install the new packages: Create a Next.js ro

Introducing Enterprise Flexible Commitment for Vercel Marketplace

Enterprise customers can now apply a portion of their Flexible Commitment toward eligible resources purchased through the Vercel Marketplace. With Flex Commit support, eligible Marketplace cost can draw directly from your existing commitment making it easier to provision the infrastructure and services your applications need. We're rolling this out with an initial set of Marketplace partners, including , , and , with support for more providers coming.NeonSupabaseRedis Eligible integrations are m

A Bootiful Podcast: Spring Boot founder and lead Phil Webb

Search encrypted data with Prisma 8 and CipherStash

CipherStash brings searchable field-level encryption to Prisma 8: encrypted equality, free-text and range queries, and identity-based key management.

Don't Let Your AI Agent Delete Your Production Database | Prisma 8

AI coding agents keep wiping production databases. Prisma 8 removes the failure modes structurally: no reset command, no shadow database, and migrations that verify every step.

Why Your RAG Pipeline is Lying to You

RAG (retrieval-augmented generation) is how most AI tools search your company's documents to answer questions. Instead of guessing, the model pulls relevant text from your files and uses it to respond Sounds reliable. Usually isn't. Last month, our team shipped a RAG-powered search feature to a Fortune 500 client. Three days later, this Slack message showed up : "The AI pulled a revenue number from our Q3 report. But when I checked the PDF, that number was from a footnote about projected revenue