SOURCE ARCHIVE · ORIGINAL LINKS INCLUDED
Vercel is reducing storage retention for Hobby projects to free up space, keeping only the three most recent deployments of each type and deleting older ones immediately if over the 10GB limit.
Mem0 launches as a native Vercel Marketplace integration, enabling AI agents and apps to retain long-term memory across user sessions with automated provisioning and billing.
Vercel’s Is Agentic now tailors audit reports by site type—Docs & content, Business, App, or Commerce—highlighting relevant checks while keeping scores consistent across views.
TypeSafe AI’s Jev probabilistic decision model is now accessible via Vercel’s AI Gateway, promising faster, cheaper, and structured decision-making for software workflows.
Delphi rebuilt its Python backend on Vercel to ship 100+ times daily, replacing a cumbersome AWS setup that slowed onboarding and deployments.
Google’s Gemini 3.8 Live and Extended Thinking models are now accessible via Vercel’s AI Gateway, enabling real-time audio interactions and parallel reasoning for voice assistants and conversational apps.
Vercel’s AI SDK harness layer now authenticates harnesses via native subscriptions when supported, eliminating code changes and enabling seamless agent switching under the same interface.
Featured, a PR co-pilot platform, migrated to Vercel to streamline infrastructure and accelerate media pitch delivery, handling over 100,000 pitches monthly with a three-person team.
Tailscale integrated Vercel AI Gateway and Sandbox to launch Aperture, a model router that centralizes AI access and agent execution within private networks, replacing fragmented provider keys with a unified, secure...
Vercel now promotes FastAPI frontends and static files to its CDN at build time, serving them directly without invoking functions unless routes or middleware require runtime checks.
Vercel reduced CDN metadata lookup latency by 91% by grouping routing paths into sharded files, cutting cache misses and speeding deployments through optimized data structures and caching strategies.
Vercel expanded Sandbox availability to all 20 compute regions, reducing latency and supporting data residency requirements with configurable failover options.
Vercel’s AI Gateway offers free access to Tako Search until September 30, enabling AI models to query curated and live web data with filtering and citations before standard billing begins.
Vercel now offers free deployment protection for all projects across Hobby, Pro, and Enterprise plans, replacing a previously paid add-on.
Vercel’s eve agents now support persistent memory, allowing context retention across sessions via configurable slots and supported providers.
Vercel’s v0 now offers one-click integrations for email, authentication, search, and databases, simplifying provider setup directly within the chat interface.
Vercel CLI now lets users read and search changelogs directly from the terminal, aiding developers and agents in tracking updates and new features.
Vercel’s Sandbox routing now uses regional replicas to resolve public domains, cutting median lookup latency from 62ms to 3.4ms and up to 146x faster in Cape Town.
Vercel introduces Flat Rate CDN for Pro teams, replacing usage-based billing with a predictable monthly fee, built-in spike protection, and premium network performance.
Vercel’s Flat Rate CDN is now generally available for Pro teams, replacing usage-based billing with a fixed monthly fee and team-level capacity.
OpenAI’s GPT Image 2.5 Flare and Sunburst models are now accessible via Vercel’s AI Gateway, offering enhanced image generation and editing capabilities with distinct performance profiles.
Vercel announced support for running Claude Managed Agents with its Chat SDK, enabling persistent, streamed agent interactions across multiple platforms without requiring server-side state management.
Vercel’s AI SDK harness layer now includes Cursor via the @ai-sdk/harness-cursor adapter, enabling developers to switch between multiple coding agents without modifying application code.
Vercel’s AI Gateway now offers Ling 3.0 Flash Fin by Inclusion AI for free until September 25, featuring a 256K-token context window and 32K output tokens for financial research.
Vercel’s AI Gateway now supports Google’s Gemini 3.5 Transcribe, offering real-time and batch audio-to-text transcription with language detection and custom vocabulary support.
Z.ai’s GLM-5.3 Flash multimodal model is now accessible via Vercel’s AI Gateway, offering a 1M token context window and support for text, vision, function calls, and streaming.
Meta’s Muse Image model is now accessible via Vercel’s AI Gateway, enabling image generation and editing through a single API endpoint with unified billing and reliability features.
Vercel’s AI Gateway now supports Alibaba’s Qwen 3.8 Flash model, enabling text and image input with a 1 million token context window and 65k token output. The model targets coding, tool use, and multi-step agent...
Vercel states its hosted Next.js applications are protected from two August 2026 security flaws without customer action, while self-hosted users must upgrade.
Vercel introduced a free tier for Speed Insights, expanding access to real-user performance data without requiring a paid plan. The free tier now persists across projects, while the paid version, Speed Insights Plus,...