Vercel 在 2026年6月 发布的内容

Pricing change detected for Vercel

Pricing updated for Vercel: - Hobby: promotion ended (was: Free forever) - Pro: promotion ended (was: $20 included credit to use across resources) - Pro: allowance changed from 1 project included to 1…

Vercel launches VCR: OCI-compliant container registry integrated with Vercel infrastructure

Vercel introduced VCR, an OCI-compliant container registry hosted on Vercel infrastructure, enabling direct push/pull operations via standard Docker commands. Images are automatically optimized for Ve…

Vercel launches Services to run full-stack applications in one project

Vercel introduced Vercel Services, allowing developers to deploy frontend, backend, and other services within a single Vercel project. Features include atomic deployments, shared preview environments,…

Vercel and Shopify rebuild Hydrogen as open-source, runtime-agnostic framework

Vercel and Shopify announced a collaboration to rebuild Hydrogen, a headless storefront framework, as an open-source, runtime-agnostic tool. The new Hydrogen centralizes core commerce logic, client-si…

Vercel enables Dockerfile deployments with Fluid compute and active CPU pricing

Vercel introduced native support for deploying any Dockerfile via a new `Dockerfile.vercel` file, allowing developers to run containers on its Fluid compute platform with automatic scaling and active …

Vercel expands Audit Log coverage via Vercel Drains for Enterprise teams

Vercel now captures 400+ team activity events in Audit Logs, replacing Custom SIEM Log Streaming with a Drains-based workflow for Enterprise teams. Audit Log Drains export events to HTTP endpoints or …

Vercel launches Services for multi-framework full-stack deployments

Vercel introduced Services, enabling developers to deploy multiple frontends and backends within a single project on a shared domain. Services communicate privately via internal bindings, deploy and r…

Vercel Sandbox adds support for custom images in public beta

Vercel Sandboxes now support custom images in public beta, allowing users to launch Sandboxes with their own root filesystem pulled from the Vercel Container Registry. This enables bringing custom OS,…

Vercel Functions now supports Dockerfile deployments for HTTP servers

Vercel Functions introduced support for deploying HTTP servers from Dockerfiles or Containerfiles using OCI-compatible images on Fluid compute. This allows existing applications in any language to lev…

Vercel Private Blob graduates to general availability with OIDC and signed URLs

Vercel Private Blob is now generally available for all plans, enabling secure storage of sensitive files with private access controls. The release introduces OIDC authentication for short-lived tokens…

Vercel expands Agent with dashboard chat, investigations, and approved actions in public beta

Vercel launched expanded Agent capabilities in public beta, integrating chat, production issue investigations, and approved actions directly into the dashboard. Agent now leverages deployment data, lo…

Vercel adds Google’s Nano Banana 2 Lite to AI Gateway with 50% cost reduction

Vercel integrated Google’s Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) into its AI Gateway, offering faster, lower-cost image generation at $0.034 per 1K images. The model supports multimodal tex…

Vercel Agent shifts from per-request to token-based pricing

Vercel Agent replaced its $0.30 per-request fee with a $0.25 per million token rate plus provider inference costs, scaling with task intensity. New users adopt the change immediately, while existing c…

Vercel adds Anthropic’s Claude Sonnet 5 to AI Gateway with launch pricing

Vercel integrated Anthropic’s Claude Sonnet 5 into its AI Gateway, offering improved coding and agentic capabilities at Sonnet pricing. Launch pricing of $2/M input and $10/M output tokens runs throug…

Vercel Open Source Program welcomes Spring 2026 cohort of 15 projects

Vercel introduced its Spring 2026 cohort of 15 open-source projects into its developer program, providing $3,600 in platform credits and third-party service credits to maintainers. The cohort includes…

Vercel Functions package size limit increased to 5GB for AI and data workloads

Vercel raised the package size limit for Node.js and Python Functions to 5GB on Fluid compute, a 20x increase from 250MB, enabling support for large AI/data libraries and heavy dependencies. The chang…

Vercel integrates xAI Grok audio models into AI Gateway

Vercel added xAI’s Grok audio models—realtime voice, text-to-speech, and speech-to-text—to its AI Gateway, enabling unified routing, observability, and spend controls via the AI SDK 7. Developers can …

Vercel adds realtime voice, TTS, and STT to AI Gateway

Vercel expanded AI Gateway to support audio/voice capabilities, including realtime voice agents, text-to-speech, and speech-to-text. The features integrate with existing AI Gateway routing, observabil…

Vercel CLI now supports querying Speed Insights metrics

Vercel introduced a new capability in its CLI allowing direct querying of Speed Insights datapoints, including core Web Vitals (LCP, INP, CLS) and other performance metrics (FCP, TTFB) from real user …

Vercel AI Gateway adds realtime voice, speech, and transcription in beta

Vercel’s AI Gateway now supports realtime voice agents, text-to-speech, and speech-to-text in beta, enabling low-latency audio interactions with the same observability and spend controls as existing t…

Pricing updated for Vercel: - Pro: now includes 1 project included - Enterprise: promotion changed — Contact our sales team to learn more

Vercel adds Agent Runs observability for eve AI agents

Vercel introduced Agent Runs in its dashboard to trace and debug eve agent sessions without OpenTelemetry setup. The feature provides detailed views of agent runs, including triggers, duration, token …

Vercel Sandboxes: Snapshots now expire based on last use

Vercel Sandbox snapshots now expire based on last use rather than creation date, with active snapshots persisting as long as workflows depend on them. Unused snapshots follow a retention policy, reset…

Vercel adds Web Analytics query support to the Vercel CLI

Vercel introduced a new `vercel metrics` CLI command enabling direct querying of Web Analytics data, including page views, visitors, and custom events. This allows developers to analyze traffic, compa…

Pricing updated for Vercel: - Pro: promotion changed — $20 included credit to use across resources - Enterprise: promotion changed — Contact sales team for custom pricing

Vercel Ship Berlin 2026: AI SDK 7 and agentic infrastructure updates

Vercel unveiled AI SDK 7, a toolkit for building and running agents with reasoning, tool calls, and multi-turn workflows, now downloaded 16M times weekly. The company also introduced Vercel Services f…

Vercel introduces product-design system for agent-native product governance

Vercel launched `product-design`, an agent-native system to embed product reasoning into their codebase for coding agents. The system includes a skill for agents, linters for enforcement, and a review…

Vercel adds Deep Agents and OpenCode adapters to AI SDK Harness

Vercel introduced two new adapters for its AI SDK Harness: Deep Agents and OpenCode, enabling developers to run these coding-agent runtimes within a unified Vercel Sandbox interface. Both adapters sup…

Vercel AI SDK 7: Production-grade agent features with MCP, tool approvals, and durability

Vercel released AI SDK 7, a major update to its TypeScript SDK for building AI agents, adding production-grade features like MCP App support, tool approvals, durable workflows via WorkflowAgent, and p…

Pricing updated for Vercel: - Pro: promotion changed — $20/month + additional usage - Pro: now includes Includes $20 monthly credit for usage - Enterprise: promotion changed — Contact sales for pricin

Pricing updated for Vercel: - Pro: promotion changed — Includes $20 monthly credit for additional usage - Enterprise: promotion changed — Contact sales for a trial and custom pricing

Vercel adds GLM 5.2 Fast via Wafer to AI Gateway with 2x throughput

Vercel’s AI Gateway now supports GLM 5.2 Fast via Wafer, delivering 2x higher throughput than other providers in benchmark tests. Small-context generation reaches 170+ tokens/sec and large-context 200…

Vercel AI SDK 7 launches with agent platform and production-grade features

Vercel released AI SDK 7, transforming its TypeScript toolkit into a full agent platform for building, running, and observing production agents across text, audio, realtime, image, and video. The upda…

Vercel Pro teams gain up to 500 concurrent builds

Vercel increased concurrent build capacity for Pro teams from 12 to 500, enabling faster deployments for large repositories. On-demand concurrent builds are now default for Pro and Enterprise plans, b…

What is v0?

Agent Stack

Vercel introduced Agent Stack, a suite of tools and templates for building AI-powered agents and Slack bots. Key components include Vercel Connect for secure third-party API access, AI SDK for agent d…

Domains

Pricing updated for Vercel: - Pro: now includes 1 developer seat included

Vercel CLI now preserves .env.local when linking

Vercel CLI v latest preserves `.env.local` files during `vercel link` operations, preventing overwrites of existing variables. The update only modifies or appends `VERCEL_OIDC_TOKEN` while leaving oth…

Vercel adds custom OIDC token audiences for secure service-to-service auth

Vercel’s OIDC issuer now supports custom audiences in tokens, allowing deployments to request provider-specific tokens for secure service-to-service authentication. The exchange service updates the au…

Vercel adds Claude Design as a send-to destination for live deployments

Vercel now integrates with Claude Design, allowing users to deploy designs directly to Vercel from the design canvas and receive a live URL. The integration requires connecting the Vercel MCP server.…

Vercel redesigns trace viewer for Workflows with local inspection

Vercel launched a redesigned trace viewer for Workflows and Workflow SDK, enabling faster inspection of runs with search, zoom, and keyboard navigation. The tool is now available locally via the Workf…

Vercel adds zero-config Node.js server deployment

Vercel now supports deploying Node.js servers with zero configuration by detecting a server.ts file at the project root or src/server.ts. This expands its zero-config backend options beyond Express, K…

Pricing updated for Vercel: - Pro: promotion changed — $20/month + additional usage - Pro: now includes Included $20 credit to use across resources - Enterprise: promotion changed — Contact sales team

Vercel CLI adds signed blob URL support for secure access control

Vercel CLI now supports generating signed URLs for Vercel Blob via the `vercel blob presign` and `vercel blob signed-token` commands. These scoped, time-limited URLs enable granular access control for…

Vercel Functions now support WebSocket connections in public beta

Vercel Functions now support WebSocket connections in public beta, enabling bidirectional realtime communication for AI streaming, chat, and collaborative apps. WebSocket invocations follow the same p…

Vercel increases projects-per-repo limit for Hobby users to 25

Vercel raised the projects-per-repository limit for Hobby users from 10 to 25, enabling easier adoption of monorepos and multi-app deployments from a single codebase. The change supports use cases whe…

Vercel Chat SDK adds Sendblue integration for iMessage, SMS, and RCS

Vercel’s Chat SDK now includes a Sendblue adapter, enabling bots to send and receive iMessage, SMS, and RCS messages via Sendblue’s gateway. The integration supports iMessage-first delivery with SMS/R…

Vercel Chat SDK adds Kapso WhatsApp integration

Vercel’s Chat SDK now includes a vendor-official adapter for Kapso, enabling WhatsApp Business integration via Kapso’s hosted platform. The adapter maps WhatsApp conversations to Chat SDK threads, sup…

Vercel Chat SDK adds Linq adapter for iMessage and group chat bots

Vercel’s Chat SDK now includes a vendor-official Linq adapter enabling bots to interact via direct messages, group chats, and iMessage tapbacks. The adapter maps Linq chats to Chat SDK threads and rea…

Vercel Chat SDK adds Novu integration for multi-channel agent support

Vercel’s Chat SDK now includes an official Novu adapter, enabling agents to operate across Slack, Teams, WhatsApp, Telegram, and email with a single handler. Novu manages credentials and delivery, whi…

Vercel AI Gateway adds Sakana Fugu Ultra model

Vercel integrated Sakana AI's Fugu Ultra model into its AI Gateway, enabling unified API access with built-in cost tracking, retries, and failover. Fugu Ultra coordinates multiple frontier models to d…

Vercel launches platform-native feature flags with server-side evaluation

Vercel introduced Vercel Flags, a platform-native feature flagging system that evaluates flags server-side by default, eliminating client-side latency and layout shifts. The system integrates directly…

Pricing updated for Vercel: - Plan features or structure changed

Pricing updated for Vercel: - Pro: promotion changed — Includes $20/month credit for additional usage - Pro: now includes 1 developer seat included - Enterprise: promotion changed — Contact sales for

Pricing updated for Vercel: - Pro: promotion changed — $20 included credit to use across resources - Enterprise: promotion changed — Contact sales team

Vercel debuts eve open source agent framework, tries to fix shadow AI with Passport

Vercel introduced eve, an open-source agent framework using TypeScript and Markdown, designed for simplicity and sandboxed execution. It includes Passport, a feature to bring AI-created employee apps …

Pricing updated for Vercel: - Pro: promotion changed — Includes $20/month credit for additional usage - Pro: now includes Pay only for active compute and set your own spend limits - Enterprise: promot

Pricing updated for Vercel: - Pro: new promotion — Includes $20 monthly credit for additional usage - Pro: annual price $20/mo → $16/mo - Enterprise: promotion changed — Contact sales for custom prici

Pricing updated for Vercel: - Pro: promotion ended (was: $20 included credit to use across resources) - Pro: annual price $16/mo → $20/mo - Pro: allowance changed from 1 developer seat included to 1 p

Vercel launches Agent Stack: AI SDK, Workflow SDK, AI Gateway, Connect, Chat SDK, and eve framework

Vercel introduced the Agent Stack, a suite of tools to build production-grade AI agents, including AI SDK for unified model access, AI Gateway for token routing, Workflow SDK for durable execution, Ve…

Pricing updated for Vercel: - Pro: promotion changed — $20 included credit to use across resources - Pro: annual price $20/mo → $16/mo - Pro: allowance changed from Includes $20 credit to use across r

Vercel Ship 2026: Agent Stack, eve framework, Vercel Connect, and Vercel Services launch

Vercel unveiled a suite of agent-focused infrastructure at Ship 2026, including the Agent Stack (AI SDK, AI Gateway, Workflow SDK, Sandbox, Chat SDK), the eve open-source agent framework, Vercel Conne…

Vercel removes CLI deployment limits for faster deployments

Vercel removed CLI-specific deployment limits, enabling faster deployments from local machines and external CI/CD pipelines with instant feedback. Teams and AI agents can now deploy at their required …

Vercel launches Vercel Connect for secure agent access to external services

Vercel introduced Vercel Connect, a system enabling apps and agents to access external services like Slack, GitHub, and Salesforce without storing long-lived secrets. It issues scoped, short-lived tok…

Vercel launches eve, an open-source agent framework

Vercel introduced eve, an open-source framework for building, running, and scaling agents with durable execution, sandboxed compute, and human-in-the-loop approvals. Agents are defined as directories …

Vercel Passport enters public beta for enterprise access control

Vercel launched Vercel Passport in public beta, enabling enterprise teams to centralize deployment access control via existing OIDC providers like Okta or Auth0. The feature costs $100 per project per…

Vercel introduced eve, an open-source agent framework designed to simplify building, running, and scaling agents by providing built-in production features like durable execution, sandboxed compute, hu

Vercel launches Vercel Connect for runtime credential exchange in Public Beta

Vercel introduced Vercel Connect, a Public Beta feature replacing long-lived provider tokens with runtime credential exchange. Agents request short-lived, scoped tokens via OIDC-verified identity, red…

Pricing updated for Vercel: - Pro: annual price $16/mo → $20/mo - Pro: allowance changed from Includes $20 credit/month for additional usage to Includes $20 credit to use across resources - Enterprise

GLM 5.2

Cancellation

Workflow DevKit v5 introduces two cancellation mechanisms: AbortSignal for fine-grained, cooperative cancellation of individual operations and run cancellation for stopping entire workflows. AbortSign…

TanStack Start

TanStack Start now supports durable workflows via the Workflow SDK, integrated as a Vite plugin. Developers can create workflows, define steps, and trigger them via API routes, with optional TypeScrip…

Vercel Workflow SDK 5 beta adds inflight cancellation with AbortController

Vercel’s Workflow SDK 5 beta now supports standard AbortController and AbortSignal APIs for in-flight cancellation across workflows and steps. Developers can pass cancellation signals into steps, enab…

Vercel launches Enterprise Apps and Agents with Passport, Connect, and BYOC

Vercel introduced Vercel for Enterprise Apps and Agents, a platform enabling secure AI agent and internal app deployment behind corporate identity and security boundaries. Key components include Verce…

Vercel Sandbox runtime extended to 24 hours for Pro and Enterprise plans

Vercel increased Sandbox runtime from 5 to 24 hours, enabling longer workloads like data processing and agentic workflows. The change applies to Pro and Enterprise plans and supports persistent sandbo…

Vercel Workflow SDK integrates with TanStack Start

Vercel’s Workflow SDK now supports TanStack Start applications, leveraging the existing `workflow/vite` plugin. Developers can add durable, resumable workflows with TypeScript using standard directive…

Vercel adds GLM 5.2 to AI Gateway with 1M token context window

Vercel’s AI Gateway now supports GLM 5.2, a model optimized for long-horizon tasks with a 1M token context window (up from 200K). It improves reliability and engineering standard adherence. Developers…

Vercel increases Blob store limit for Hobby users to 100

Vercel raised the Blob store limit for Hobby users from 5 to 100, enabling better data organization by project, environment, or region. Storage, operations, and transfer limits remain unchanged.

Vercel Functions extend max runtime to 30 minutes for Pro and Enterprise

Vercel increased the maximum execution duration for Node.js and Python Functions to 30 minutes for Pro and Enterprise plans, up from 800 seconds. This enables longer AI reasoning, document processing,…

Pricing updated for Vercel: - Pro: promotion changed — $20/month + additional usage - Pro: allowance changed from 1 project included to Includes $20 credit/month for additional usage - Enterprise: pro

Pricing updated for Vercel: - Pro: promotion changed — Includes $20/month credit for additional usage - Pro: allowance changed from 1 developer seat included to 1 project included - Enterprise: promot

Pricing updated for Vercel: - Pro: now includes 1 developer seat included - Enterprise: promotion changed — Contact sales team for custom pricing

Deploy a Claude Design project with Vercel Drop

Vercel Drop enables deploying Claude Design projects by dragging a .zip export directly into the browser, eliminating the need for Git or CLI. This streamlines the workflow from prompt-based design to…

Deploy a Google Stitch design with Vercel Drop

Vercel introduced an integration with Google Stitch, allowing users to export HTML designs from Stitch and deploy them instantly via Vercel Drop without Git, CLI, or local setup. The workflow streamli…

Deploy a Bolt.new app with Vercel Drop

Vercel introduced support for deploying Bolt.new projects via Vercel Drop, allowing users to export a Bolt project as a .zip and drag it into Vercel for instant deployment. Vercel automatically detect…

Nitro

The Workflow SDK's Nitro v3 module now includes a new setup guide for creating workflows, requiring users to add the 'workflow/nitro' module to their Nitro config and use TypeScript directives like 'u…

Workflow SDK now runs natively in Vercel Nitro v3

Vercel’s Workflow SDK now integrates natively with Nitro v3, eliminating separate bundling by running steps inside the app’s runtime. This enables direct use of Nitro’s server-side APIs like `useStora…

Kimi K2.7 Code

Vercel’s AI Gateway now includes Kimi-K2.7-Code, a coding model from Moonshot AI, featuring enhanced coding and agent performance, reduced overthinking, and better instruction following for long-horiz…

AI SDK

AI SDK Harnesses

AI SDK v7 adds a harness abstraction to run established agent runtimes like Claude Code or Codex through a unified interface, offering workspace access, tools, and session state while operating in a s…

Statement on the US government directive to suspend access to Fable 5 and Mythos 5

The US government ordered Anthropic to suspend all access to Fable 5 and Mythos 5 for foreign nationals, citing a potential jailbreak method. Anthropic complied but disputes the severity, arguing the …

Pricing updated for Vercel: - Enterprise: promotion changed — Contact sales team

Vercel adds Moonshot AI's Kimi K2.7 Code to AI Gateway

Vercel integrated Moonshot AI's K2.7 Code model into its AI Gateway, enabling developers to access the multimodal coding model via a unified API with built-in usage tracking, cost management, and fail…

Vercel AI SDK 7 introduces HarnessAgent for unified agent harnesses

Vercel’s AI SDK 7 adds `HarnessAgent`, a single API to run agent harnesses like Claude Code, Codex, and Pi without rewriting code. Developers can switch harnesses dynamically while maintaining the sam…

Vercel suspends Claude Fable 5 access on AI Gateway due to US government directive

Vercel suspended access to Anthropic’s Claude Fable 5 model on its AI Gateway platform, effective immediately, following a US government directive. The suspension applies to all users, with no confirm…

Vercel Drop enables zero-setup deployments via drag-and-drop

Vercel launched Drop, a browser-based drag-and-drop deployment tool that publishes files or folders to production instantly without Git, CLI, or local setup. It supports static sites and framework pro…

Pricing updated for Vercel: - Pro: promotion changed — $20 included credit to use across resources - Pro: allowance changed from Includes $20 credit/month for usage to 1 developer seat included - Ente

Pricing updated for Vercel: - Pro: promotion changed — $20/month + additional usage - Pro: allowance changed from 1 developer seat included to Includes $20 credit/month for usage

Vercel: Okara integrates AI Gateway and Sandboxes for 120K companies

Okara, an AI CMO serving 120,000+ companies, adopted Vercel AI Gateway to unify eight AI model providers under one API key, eliminating per-provider SDKs and reducing infrastructure overhead. They als…

Vercel AI Gateway adds Azure as provider for DeepSeek V4 Pro and V4 Flash

Vercel’s AI Gateway now supports DeepSeek V4 Pro and V4 Flash via Azure, enabling failover routing and optional prioritization of Azure requests without code changes. Existing Azure credentials can be…

Vercel plugin now available in Grok Build

Vercel announced its plugin is now integrated with Grok Build, enabling real-time knowledge injection for Vercel APIs and patterns during coding sessions. Users can install it via prompt recommendatio…

Vercel: Okara integrates AI Gateway and Sandboxes for AI CMO platform

Okara, an AI CMO platform, now runs its multi-provider AI stack on Vercel, processing 4B tokens daily for 120,000+ businesses. Vercel AI Gateway replaced eight separate SDKs with a single endpoint, en…

Pricing updated for Vercel: - Pro: promotion changed — $20 included credit to use across resources - Enterprise: promotion changed — Contact sales for a trial and custom pricing

Vercel enables threshold billing for Pro teams

Vercel introduced threshold billing for Pro teams, issuing partial invoices mid-cycle once on-demand usage hits a set threshold instead of waiting until the end of the billing period. Total usage rema…

Pricing updated for Vercel: - Pro: promotion changed — Includes $20/month credit for additional usage - Pro: allowance changed from Includes $20 credit for additional usage to 1 developer seat include

Claude Fable 5

Pricing updated for Vercel: - Pro: promotion changed — $20/month + additional usage - Pro: now includes Includes $20 credit for additional usage - Enterprise: promotion changed — Contact sales for a t

Code and Theory cuts time-to-prototype 75% using Vercel's v0

Code and Theory, a digital agency, reduced time-to-prototype by 75% and deployment timelines by over 50% using Vercel's v0 prompt-to-code workflows. The agency replaced traditional wireframes and PRDs…

The Weather Company serves 350M users on Vercel with 80% velocity boost

The Weather Company rebuilt its web serving stack on Vercel, serving 350M monthly active users and processing 2.2B coordinates every 15 minutes. The migration reduced time from design to live page by …

Vercel adds spend quotas for AI Gateway API keys

Vercel introduced spend quotas for AI Gateway API keys to cap costs from autonomous workflows, demos, or unchecked experimentation. Users can set dollar limits and refresh periods (daily, weekly, mont…

Vercel CLI adds domain search with TLD filtering and sorting

Vercel introduced a new `vercel domains search` CLI command to check domain availability and pricing across supported TLDs, with filtering, sorting, and JSON output options. Users must upgrade to CLI …

Vercel adds Anthropic’s Claude Fable 5 to AI Gateway

Vercel’s AI Gateway now supports Anthropic’s Claude Fable 5, a Mythos-class model optimized for long-running, multi-step tasks with improved code review and first-shot correctness. The model includes …

Fern migrates docs to Vercel, cutting load times by 80%

Fern, a docs and SDK platform, migrated its multi-tenant documentation to Vercel, reducing page load times by 80% and achieving 3x faster time to first byte. The team moved 65% of its platform from Pa…

Pricing updated for Vercel: - Pro: promotion changed — Includes $20 monthly credit for additional usage - Enterprise: promotion changed — Contact sales team for custom pricing

Vercel AI Gateway Index: DeepSeek V4 disrupts AI model usage in production

Vercel’s AI Gateway data for May 2026 shows DeepSeek V4’s token share surged to 17% in a month while spend share remained near 1%, disrupting production AI workflows. Anthropic maintained 65% of spend…

Pricing updated for Vercel: - Pro: promotion changed — $20/month + additional usage - Pro: allowance changed from 1 developer seat included to Includes $20 monthly credit for usage - Enterprise: promo

Pricing updated for Vercel: - Pro: promotion changed — $20 included credit to use across resources - Pro: allowance changed from Includes $20 monthly credit to 1 developer seat included - Enterprise:

Pricing updated for Vercel: - Pro: promotion changed — $20/month + additional usage - Pro: allowance changed from 1 developer seat included to Includes $20 monthly credit - Enterprise: promotion chang

Pricing updated for Vercel: - Pro: promotion changed — + $20 included credit to use across resources - Pro: allowance changed from Includes $20 credit to use across resources to 1 developer seat inclu

Shopify Online Store with Next.js

Vercel introduced a new Next.js 16 template for building Shopify online stores, featuring React 19, Tailwind CSS 4, and the Shopify Storefront API. The template includes a Vercel Shop plugin with comm…

Pricing updated for Vercel: - Pro: promotion changed — $20/month + additional usage - Pro: allowance changed from 1 developer seat included to Includes $20 credit to use across resources - Enterprise:

Vercel Sandbox adds persistent drives in private beta

Vercel Sandbox now supports persistent drives in private beta, allowing users to create and mount storage that persists beyond sandbox lifecycles. This enables retaining workspaces, repositories, and …

Vercel integrates skills.sh API with OIDC token authentication

Vercel launched the skills.sh API, enabling teams to query over 600,000 open-source skills using short-lived Vercel OIDC tokens. The API supports searching, detailed lookups, and security audits with …

Nemotron 3 Ultra

Pricing updated for Vercel: - Enterprise: promotion changed — Contact sales team for custom pricing

Pricing updated for Vercel: - Pro: allowance changed from 1 Developer seat included to 1 developer seat included - Enterprise: promotion changed — Contact sales for a trial and custom pricing

Vercel AI Gateway adds Nvidia Nemotron 3 Ultra for agentic workflows

Vercel integrated Nvidia’s Nemotron 3 Ultra, an open Mixture-of-Experts model optimized for long-running agent workflows, into its AI Gateway. The model offers a 1M token context window, 350 tokens/se…

Vercel updates legal terms to address AI functionality and billing flexibility

Vercel introduced new terms clarifying responsibilities for AI-powered services (AI Functionality) and third-party tools accessing accounts, alongside billing changes for non-subscription services and…

Vercel launches Shopify integration for storefront deployment

Vercel introduced a new Shopify integration enabling users to create and deploy Shopify storefronts directly from its platform in minutes. The integration automates credential configuration, supports …

Grok Imagine Video 1.5 Preview

Vercel’s AI Gateway now supports xai/grok-imagine-video-1.5-preview, enabling users to generate videos from text prompts via an experimental API. The model is priced at $0.01 per image and is availabl…

Pricing updated for Vercel: - Pro: promotion changed — $20 included credit to use across resources - Pro: allowance changed from Includes $20 monthly credit for usage to 1 Developer seat included - En

Vercel CLI adds --trace flag for OpenTelemetry request tracing

Vercel introduced a new `vercel curl --trace` CLI command to generate OpenTelemetry traces for any request from the terminal, along with a `vercel traces get` command to fetch traces by request ID. Th…

Vercel adds xAI’s Grok Imagine Video 1.5 to AI Gateway

Vercel integrated xAI’s Grok Imagine Video 1.5 into its AI Gateway, enabling single-pass video generation from images with synchronized audio. The update improves audio quality, photorealism, and prom…

Pricing updated for Vercel: - Pro: new promotion — $20 included credit to use across resources - Enterprise: promotion changed — Contact sales for a trial and custom pricing

Chat SDK launches official Lark/Feishu adapter with WebSocket and card streaming

The Chat SDK team released an official adapter for Lark/Feishu, enabling WebSocket long-connection event delivery, native cardkit typewriter streaming, interactive cards, and reactions. The adapter us…

Vercel enables bulk Git settings editing for monorepos

Vercel introduced a feature allowing users to edit Git settings for all projects within a monorepo in a single place, eliminating the need to configure each project individually. This change simplifie…

Vercel Blob now supports time-bound signed URLs for secure uploads and downloads

Vercel introduced signed URLs for Vercel Blob, enabling time-bound, operation-scoped access (up to 7 days) for single operations like upload, download, or delete. These URLs support direct browser upl…

Qwen 3.7 Plus

Pricing updated for Vercel: - Pro: promotion ended (was: Includes $20 monthly credit for additional usage) - Pro: now includes 1 developer seat included - Enterprise: promotion changed — Contact sales

Pricing updated for Vercel: - Pro: promotion changed — Includes $20 monthly credit for additional usage - Enterprise: promotion changed — Contact sales for custom pricing

MiniMax M3

Vercel Blob pricing structure overhaul with regional rates and Hobby plan limits

Vercel introduced a new regional pricing model for Blob storage with separate charges for storage, operations, and data transfer. Hobby users now face usage limits with no overage fees but temporary a…

Vercel Blob switches to OIDC authentication by default

Vercel Blob now defaults to OIDC authentication, replacing long-lived tokens with short-lived, auto-rotating tokens. Existing stores require manual upgrade via the Projects tab, while new projects use…

Vercel adds elastic build machines to prevent out-of-memory failures

Vercel introduced elastic build machines that automatically adjust memory tiers to prevent out-of-memory (OOM) build failures. The system upgrades or avoids downgrades based on memory usage, improving…

Vercel AI Gateway adds Alibaba's Qwen 3.7 Plus model

Vercel’s AI Gateway now supports Alibaba’s Qwen 3.7 Plus, a multimodal agent foundation model unifying vision and language. Users can access it via the AI SDK with the model identifier `alibaba/qwen3.…