- vercel.com
Vercel publishes guide on AWS App Runner alternatives, highlighting Container Runtime and Services
Vercel published a detailed guide comparing AWS App Runner alternatives, emphasizing Vercel’s Container Runtime (Beta) for HTTP applications with frontends and Vercel Services (Beta) for bundling fron…
- Contentvercel.com
Vercel publishes guide on DigitalOcean App Platform alternatives
Vercel released a detailed comparison guide highlighting when Vercel Services, Container Runtime, and Daemons are suitable replacements for DigitalOcean App Platform. The guide positions Vercel as a f…
- Featurevercel.com
Vercel adds Ling 3.1 Flash 1019 to AI Gateway with free access until November 3, 2026
Vercel integrated Ling 3.1 Flash 1019 from InclusionAI into its AI Gateway, offering free access to the model until November 3, 2026. The model supports coding, multi-step reasoning, and agent workflo…
- Featurevercel.com
Vercel adds HTTP liveness checks to Daemons for automatic restarts
Vercel Daemons now support HTTP liveness checks via a new `livenessProbe` configuration in vercel.json. Users can define a lightweight health endpoint in their daemon code, and Vercel will automatical…
- Featurevercel.com
Vercel Daemons gain service bindings for internal HTTP calls
Vercel Daemons can now declare service bindings to call other services in the same deployment via environment variables, with Vercel handling routing, authentication, and TLS. Bindings are one-way (se…
- Launchvercel.com
Vercel launches Geist Serif as new display serif font in the Geist family
Vercel introduced Geist Serif, a high-contrast display serif font designed for modern screens and print, expanding its Geist typeface family. The font features saw-toothed serifs, sharp hairlines, and…
- Featurevercel.com
Vercel migrates Data Cache to Runtime Cache with new pricing and storage tiers
Vercel is replacing Data Cache with Runtime Cache, which will auto-enable for new teams and migrate existing Hobby, Pro, and Enterprise teams by November 29, 2026. Runtime Cache offers 8–10x more stor…
- vercel.com
Vercel consolidates Data Cache into Runtime Cache with expanded storage and performance improvements
Vercel merged its Data Cache into Runtime Cache, unifying caching APIs for frameworks and APIs without requiring code changes for Next.js `fetch` or `unstable_cache` users. Pro and Enterprise projects…
- vercel.com
Vercel Plugin now available in Google Antigravity marketplace
Vercel announced the availability of its official plugin in Google Antigravity, enabling Antigravity agents to interact with Vercel APIs and recommended patterns. The plugin supports Vercel AI SDK, Fu…
- Featurevercel.com
Vercel Flags SDK v1.5.3 improves cold-start performance and removes function duration extension
Vercel Flags SDK v1.5.3 reduces cold-start initialization time by 7x (from >70ms to ~10ms) and eliminates function duration extension by moving usage reporting to the Vercel runtime. Flag reads remain…
- vercel.com
Vercel AI Gateway now supports OpenAI API credits via BYOK for eligible ChatGPT subscribers
Vercel AI Gateway now allows eligible ChatGPT Pro and Business subscribers to claim and use their OpenAI API credits for app requests via Bring Your Own Key (BYOK). Credits are pooled per workspace an…
- Integrationvercel.com
Vercel AI Gateway now supports Claude API credits via Anthropic BYOK
Vercel AI Gateway now allows users to claim and apply Claude Max or Team API credits toward Anthropic model usage through BYOK. Eligible subscribers can link their Claude Console organization, add an …
- vercel.com
Vercel publishes guide on Fly.io alternatives for web apps and workers
Vercel released a detailed guide comparing Fly.io alternatives, explicitly positioning its platform for web apps and continuous workers. The post highlights Vercel’s Container Runtime (Beta) and Daemo…
- vercel.com
Vercel publishes guide comparing Coolify alternatives for deployment
Vercel published a detailed guide comparing Coolify alternatives, highlighting its own managed hosting for web apps and workers as a primary option. The guide contrasts Vercel, Render, Railway, and se…
- Contentvercel.com
Vercel publishes guide comparing Render alternatives for web apps and workers
Vercel released a detailed guide positioning its platform as a Render alternative, highlighting framework-based deployments, container runtime (beta), and Daemons (beta) for continuous workers. The po…
- vercel.com
Vercel publishes Docker VPS migration checklist to guide app moves to its platform
Vercel released a step-by-step guide for migrating Docker applications from VPS environments to its platform, emphasizing inventorying components, mapping to Vercel’s Container Runtime or Daemons, and…
- Integrationvercel.com
Strix integrates with Vercel AI Gateway for unified LLM access
Vercel AI Gateway now supports Strix, an open-source AI penetration testing tool, enabling it to access 400+ models from providers like Anthropic and OpenAI through a single API. Usage and costs are t…
- Featurevercel.com
Vercel adds AI Bots managed ruleset to Firewall for granular AI crawler and assistant control
Vercel introduced a new AI Bots managed ruleset in Vercel Firewall, allowing users to separately log or block AI crawlers (content collectors) and AI assistants (user-query responders). The feature is…
- Featurevercel.com
Vercel makes Node.js 26 LTS generally available for Functions and builds
Vercel now supports Node.js 26 LTS by default for all new projects and deployments, replacing prior versions. Key additions include HTTP/2 by default for fetch, the Temporal API for immutable time han…
- Featurevercel.com
Vercel Connect adds webhook triggers for Resend, X, Twilio, Notion, and Telegram
Vercel Connect now supports webhook triggers for five new providers—Resend, X, Twilio, Notion, and Telegram—expanding its real-time event forwarding capabilities. The feature automates webhook registr…
- Featurevercel.com
Vercel improves IP geolocation resolution to city-level for nearly all traffic
Vercel enhanced its IP geolocation to provide city-level data for nearly all requests and postal-code data for 90% of traffic, with more frequent updates to reflect real-time network changes. The chan…
- Pricingvercel.com
Vercel lowers Mistral Large 4 pricing on AI Gateway with updated weights
Vercel updated Mistral Large 4 weights on AI Gateway, improving agentic coding, cybersecurity, and token efficiency while reducing input, cached input, and output token pricing. Existing model IDs aut…
- Featurevercel.com
Vercel Container Registry increases compressed image layer limit to 2 GB
Vercel Container Registry now supports compressed image layers up to 2 GB each, quadrupling the previous 500 MB limit while keeping the total image size cap at 15 GB. This enables larger dependency se…
- Integrationvercel.com
Vercel Sandbox gains Box Mount integration for filesystem access
Vercel introduced Box Mount in private preview, enabling agents in Vercel Sandbox to read and write Box files via standard filesystem operations. The integration syncs a Box folder with a sandbox dire…
- Integrationvercel.com
Vercel integrates Paseo agents with Vercel Sandbox
Vercel launched a plugin enabling Paseo agents (fx, Codex, Claude Code, OpenCode, Pi) to run in the Vercel Sandbox, allowing cross-device session continuity and persistent file/conversation history. U…
- Featurevercel.com
Vercel updates serverless Python guidance with Fluid compute, Large Functions beta, and Active CPU pricing
Vercel published a comprehensive guide on serverless Python, detailing how its Fluid compute model reduces cold starts, introduces Large Functions beta (5 GB package limit), and implements Active CPU …
- Integrationvercel.com
Vercel adds Particle-1 to AI Gateway for software engineering tasks
Vercel integrated Particle-1, a post-trained AI model optimized for software engineering tasks, into its AI Gateway. The model reduces token usage and improves debugging efficiency via a 1M-token cont…
- Featurevercel.com
Vercel AI Gateway adds virtual models for centralized AI agent configuration
Vercel introduced virtual models in AI Gateway, allowing centralized configuration of AI models and settings via custom IDs like `vmc/coding-agent`. Developers can adjust reasoning effort, routing, co…
- vercel.com
Vercel AI Gateway adds Virtual Models for centralized model governance
Vercel introduced Virtual Models in AI Gateway, allowing teams to abstract model selection and configuration behind a single ID. This enables centralized control over model switching, fallbacks, Zero …
- Integrationvercel.com
Vercel integrates fx coding agent into T3 Code control plane
Vercel now supports fx, its open-source coding agent, as a provider within T3 Code, enabling Supervised and Auto modes for coding sessions across web, desktop, and mobile. Users can install fx and con…
- vercel.com
Vercel launches Vercel CI in Beta with task-level optimization
Vercel introduced Vercel CI in Beta, a new CI system that runs only tasks affected by code changes and scales compute dynamically using Turborepo and Hive. At Vercel, it reduced CI time from push to c…
- Featurevercel.com
Vercel CI enters beta for Turborepo with one-line setup
Vercel launched Vercel CI in beta, enabling one-line configuration for Turborepo to run CI tasks on every push. The feature reduces CI wait times by 2-6x in Vercel’s internal repos by running only aff…
- Featurevercel.com
Vercel CLI adds non-interactive enablement for Web Analytics and Observability Plus
Vercel CLI now allows agents to enable Web Analytics and Observability Plus programmatically via new commands (`vercel project web-analytics enable` and `vercel project observability enable`) without …
- Featurevercel.com
Vercel HarnessAgent now supports image inputs across all adapters
Vercel’s AI SDK HarnessAgent now accepts image inputs in the standard AI SDK message format, enabling vision model processing across all harness adapters including Claude Code, Cline, and GitHub Copil…
- Featurevercel.com
Vercel adds AI traffic insights to Observability with CLI support
Vercel introduced AI traffic insights in Observability to track verified AI bot requests, most-crawled pages, assistant retrievals, Markdown requests, and errors. Users can view these metrics in the d…
- Featurevercel.com
Vercel updates fx harness adapter with host-process execution and new AI features
Vercel’s fx adapter for the AI SDK’s HarnessAgent now runs the harness and model requests in the host process while keeping filesystem and shell operations isolated in the sandbox. The update adds too…
- Pricingvercel.com
Vercel Agent sessions cost 67% less after efficiency improvements
Vercel reduced Agent session costs by 67% by overhauling context management, prompt caching, and workflow-based CLI skills, making sessions more efficient without changing billing rates. Changes are l…
- Integrationvercel.com
Vercel Plugin now available in OpenCode
Vercel launched its official plugin in OpenCode, providing AI-driven guidance on Vercel APIs, deployment, performance optimization, and AI app design. Users can deploy projects and manage environment …
- Featurevercel.com
Vercel adds plain Dockerfile deployment for Container runtime
Vercel now allows deploying a plain root Dockerfile or Containerfile on its Container runtime when no framework is detected, automatically selecting the Container preset and routing traffic to the bui…
- Integrationvercel.com
Vercel integrates ElevenLabs audio models into AI Gateway with 50% discount
Vercel partnered with ElevenLabs to add 14 of its audio models to Vercel’s AI Gateway, enabling speech-to-text, streaming text-to-speech, realtime voice agents, music generation, sound effects, and mu…
- Featurevercel.com
Vercel Sandbox adds stdin support for real-time command input
Vercel Sandbox now supports sending input to running commands via stdin, enabling real-time interaction with stdio MCP servers and CLI programs. The feature is available in both TypeScript and Python …
- Featurevercel.com
Vercel CLI adds 1000+ service discovery for Vercel Connect
Vercel introduced a new CLI command, `vercel connect discover`, enabling users to search, filter, and select from over 1000 services in Vercel Connect directly from the terminal. The feature supports …
- Launchvercel.com
Vercel adds OpenAI’s GPT-6.1 Terra to AI Gateway for coding and visual reasoning
Vercel now supports OpenAI’s GPT-6.1 Terra model on its AI Gateway, targeting high-volume visual reasoning, computer use, and writing tasks. The model offers a 1M+ token context window, structured out…
- Featurevercel.com
Vercel adds customizable widgets for project cards in dashboard
Vercel’s Overview tab now lets teams replace the fixed Real Experience Score on project cards with customizable widgets, including firewall activity, observability metrics, Speed Insights, and Web Ana…
- Featurevercel.com
Vercel adds custom monitors to Observability for Pro and Enterprise plans
Vercel introduced custom monitors in Observability to alert users based on query results, supporting metrics like AI Gateway costs, Core Web Vitals, and failure rates. Users can set anomaly detection …
- Featurevercel.com
Vercel Agent adds alert validation to reduce paging noise
Vercel Agent now validates alerts from Vercel monitors before sending them to paging systems, forwarding only Critical alerts to destinations like incident.io. The feature uses log and metric analysis…
- Integrationvercel.com
Vercel Sandbox now supports Scale AI’s AgentEnv framework
Vercel Sandbox now integrates Scale AI’s AgentEnv framework (v0.9.1334+) to run and score AI agent evaluations in isolated Sandboxes. Users can deploy and test agents against tasks directly on Vercel,…
- Featurevercel.com
Vercel speeds up pnpm installs with dependency verification reuse
Vercel reduced pnpm install times by up to 14% for Pro deployments with restored build cache by reusing successful dependency verification across deployments. In best cases, unchanged dependencies now…
- Integrationvercel.com
Vercel adds Moonshot AI's Kimi K3.1 Pro models to AI Gateway
Vercel integrated Moonshot AI’s Kimi K3.1 Pro and K3.1 Pro Highspeed models into its AI Gateway, enabling developers to access these models via Vercel’s unified API. The models support advanced coding…
- Featurevercel.com
Vercel MCP adds submit_feedback tool for direct bug and feature reports
Vercel MCP now includes a `submit_feedback` tool enabling agents to send bug reports, friction reports, and feature requests directly to the Vercel team with structured context. The tool requires sign…
- Featurevercel.com
Vercel Agent introduces plan-based permission approval for safer production access
Vercel Agent now requires users to approve detailed plans outlining proposed actions and required permissions before executing production changes. Plans specify intent, work tasks, and scopes, with ap…
- Case Studyvercel.com
Interfaze serves 50M requests/month on Vercel with AI Gateway and Fluid compute
Interfaze, a deterministic AI model provider, runs its entire backend on Vercel, handling 50 million requests monthly. The setup leverages Vercel’s AI Gateway for reliable model routing, Fluid compute…
- Integrationvercel.com
Vercel Connect adds managed Stripe API keys for granular access control
Vercel Connect now supports creating restricted Stripe API keys for individual agents or tasks, replacing the default full-access secret key approach. Each key is scoped to specific permissions and ma…
- Featurevercel.com
Vercel CLI adds Spend Management budget controls
Vercel’s CLI now supports managing Spend Management budgets directly via `vercel budgets`, allowing teams to view, set, remove, and monitor budget thresholds and pause production deployments at 100% s…
- Featurevercel.com
Vercel deployments now display build cache size and limits
Vercel added build cache size visibility to deployment summaries, CLI (v65.1.0+), and the REST API, showing saved space and proximity to limits. Deployments exceeding cache limits lose cache for subse…
- vercel.com
Vercel CLI adds vercel feedback command for structured bug and feature reports
Vercel introduced a new `vercel feedback` CLI command that lets users and coding agents send structured bug reports, friction logs, and feature requests directly to the Vercel team from the terminal. …
- Contentvercel.com
Vercel explains WSGI vs ASGI for Python teams
Vercel published a technical guide comparing WSGI and ASGI for Python applications, detailing callable contracts, concurrency models, streaming, and WebSocket support. The post clarifies when switchin…
- Featurevercel.com
Vercel expands OIDC support and sets new authentication requirements for Marketplace providers
Vercel is pushing to eliminate long-lived static credentials by expanding OIDC (OpenID Connect) support across its platform and setting new authentication requirements for Marketplace providers. By No…
- Featurevercel.com
Vercel CLI enables account-free deployments with temporary URLs
Vercel introduced a new CLI feature allowing users to deploy projects without an account, generating temporary URLs via `vercel deploy --temporary` that expire after one hour unless claimed. This acce…
- Featurevercel.com
Vercel expands free domain TLDs to eight new extensions for Pro teams
Vercel added eight new TLDs (.co, .studio, .company, .info, .pro, .live, .digital, .agency) to its free domain offer for Pro teams, increasing eligible extensions to sixteen. The offer remains one dom…
- Contentvercel.com
Vercel publishes Railway migration checklist to guide workload moves
Vercel released a detailed migration checklist to help users move applications off Railway, covering service dependencies, data transfers, testing, and rollback planning. The guide emphasizes incremen…
- Featurevercel.com
Vercel removes skipped monorepo deployments from rate limits
Vercel now excludes automatically skipped monorepo deployments from deployment rate limits, allowing teams to utilize their limits more efficiently. Skipping is determined via Turborepo's package grap…
- Featurevercel.com
Vercel adds Python support to routing middleware
Vercel now supports writing routing middleware in Python for frameworks like FastAPI, Flask, and Django. The middleware runs before requests reach the app and can rewrite, redirect, or respond to them…
- Featurevercel.com
Vercel offers free Baseten decision models Mammoth-1 and Elephant-1 on AI Gateway
Vercel added Baseten’s Mammoth-1 (27B) and Elephant-1 (9B) decision models to AI Gateway for free through October 27, enabling text classification, routing, and scoring with 16k-token context windows.…
- Featurevercel.com
Vercel Functions WebSocket support now generally available
Vercel announced general availability of WebSocket support in Vercel Functions, following a public beta that served 300 million connections (15M daily). WebSockets power AI chat, streaming, multiplaye…
- Featurevercel.com
Vercel adds DeepSeek V4.1 Flash to Token Pass AI Gateway
Vercel introduced DeepSeek V4.1 Flash in its Token Pass subscription, a $30/month Pro team offering with $80 monthly credit via AI Gateway. The model is accessible through preferred providers with Zer…
- Integrationvercel.com
Vercel adds MiniMax M3.1 Flash to AI Gateway for free until Oct 20
Vercel integrated MiniMax M3.1 Flash, an open-weight multimodal model optimized for coding and agentic tasks, into its AI Gateway with a free tier until October 20. The model supports 1M-token context…
- vercel.com
Vercel AI Gateway adds Mercury 3 with 50% launch discount
Vercel AI Gateway now supports Mercury 3 from Inception, offering 50% off launch pricing and generation speeds over 1,000 tokens per second. The model uses diffusion for parallel text refinement, idea…
- Legalvercel.com
Vercel updates Hobby and Pro Terms of Service with new credit and cancellation rules
Vercel revised its Terms of Service for Hobby and Pro plans, effective October 19, 2026, clarifying credit expiration (plan credits at term end, purchased credits after one year), managed account poli…
- Featurevercel.com
Vercel explains code interpreter tooling and Vercel Sandbox enhancements
Vercel published a detailed guide explaining how code interpreters enable LLMs to write, run, and validate code in isolated environments, with a focus on differences between hosted interpreters and cu…
- Technicalvercel.com
Vercel explains Kata Containers for secure AI agent isolation
Vercel published a technical guide on Kata Containers, an open-source runtime that runs each pod in a lightweight VM with a separate kernel for stronger isolation. It contrasts Kata with Docker, gViso…
- Featurevercel.com
Vercel introduces cloud agents for remote, isolated AI coding tasks
Vercel launched cloud agents that execute coding tasks in remote, isolated environments, enabling unattended work like bug fixes, migrations, and dependency updates. These agents run in vendor-managed…
- Contentvercel.com
Vercel compares E2B alternatives, highlighting Vercel Sandbox for agent code execution
Vercel published a comprehensive guide comparing six E2B alternatives for agent-generated code execution, emphasizing isolation, state persistence, session limits, and BYOC support. The post positions…
- Featurevercel.com
Vercel launches durable-harness for resilient agent workflows
Vercel introduced `durable-harness`, an open-source tool enabling any coding agent harness (Pi, OpenCode, Codex, Claude Code) to run durably in the cloud by surviving crashes and resuming interrupted …
- Featurevercel.com
Vercel AI Gateway adds session affinity logs for cost and cache insights
Vercel’s AI Gateway now groups requests into sessions in logs, showing how session affinity routed each request and revealing cache hit rates, costs, and provider details per session. Developers can f…
- vercel.com
Vercel adds DeepSeek V4.1 Flash 1017 to AI Gateway with improved reasoning and coding
Vercel’s AI Gateway now supports DeepSeek V4.1 Flash 1017, a minor update improving reasoning, coding, agentic tasks, and image understanding. The model delivers faster task completion with shorter ou…
- Integrationvercel.com
Vercel launches Vercel Shop with Shopify’s Hydrogen SDK for agent-first storefronts
Vercel and Shopify released Vercel Shop, a generally available Next.js storefront template optimized for agents, search engines, and humans. It leverages Shopify’s rebuilt Hydrogen SDK and the Univers…
- Integrationvercel.com
Vercel adds Perplexity Decider V1.2 models to AI Gateway
Vercel integrated Perplexity’s Decider V1.2 27B and Decider Flash V1.2 4B models into its AI Gateway, enabling multimodal decision-making for routing, classification, and scoring tasks. The models sup…
- Integrationvercel.com
Vercel adds Reflection AI's Beam model to AI Gateway for free during launch
Vercel integrated Reflection AI's Beam, a low-latency open-weight reasoning model for coding and agentic tasks, into its AI Gateway. Beam is free to use during the launch period via standard and '-fre…
- Integrationvercel.com
Vercel adds Elorian AI's Glimpse visual reasoning model to AI Gateway
Vercel integrated Elorian AI's Glimpse model into AI Gateway, enabling visual reasoning with images and video for tasks like scientific chart analysis and robotics annotation. Developers can now use t…
- vercel.com
Vercel launches Drives for persistent storage in Vercel Sandbox
Vercel introduced Vercel Drives, a persistent storage solution for agents running in Vercel Sandbox, now generally available. Drives allow agents to maintain file state across different environments a…
- vercel.com
Vercel launches Python general availability with full-stack support and AI SDKs
Vercel announced Python is now generally available, enabling developers to build and run Python applications with familiar frameworks like FastAPI, Flask, and Django. The platform automatically detect…
- Featurevercel.com
Vercel expands Enterprise Apps and Agents with BYOC on AWS and Enterprise-Managed Authorization
Vercel introduced new enterprise governance and infrastructure features, including generally available Bring Your Own Cloud (BYOC) on AWS and public beta Enterprise-Managed Authorization via Vercel Co…
- Featurevercel.com
Vercel expands backend workloads: Python GA, containers, and Daemons in Public Beta
Vercel announced Python backend support is now generally available, enabling frameworks like FastAPI, Flask, and Django to run on Vercel with full infrastructure handling. Container support for Functi…
- vercel.com
Vercel adds Liquid AI's d1-ultra decision model to AI Gateway
Vercel integrated Liquid AI's d1-ultra multimodal decision model into its AI Gateway, enabling structured decision-making with calibrated probabilities for text and image inputs. The model supports Bo…
- Launchvercel.com
Vercel adds Meta’s Muse Spark Open to AI Gateway for coding workflows
Vercel integrated Meta’s Muse Spark Open, an open-weight reasoning model optimized for coding and agentic workflows, into its AI Gateway. The model supports 1M-token context windows, text/image inputs…
- Featurevercel.com
Vercel Connect adds Secure Compute routing for IP allowlisted providers
Vercel Connect now supports Secure Compute routing, enabling connectors to reach providers with IP allowlists like Salesforce and Snowflake using dedicated static IPs instead of shared ones. This feat…
- vercel.com
Vercel adds Cat Jam reasoning model to AI Gateway for free during stealth period
Vercel’s AI Gateway now offers Cat Jam, a low-latency reasoning model for coding and agentic tasks, available for free during its stealth period. The model supports function calling, structured JSON o…
- Launchvercel.com
Vercel Drives now generally available for Vercel Sandbox
Vercel Drives, a persistent storage solution for Vercel Sandbox, has reached general availability across Hobby, Pro, and Enterprise plans. Drives retain files between sandbox sessions and can be mount…
- Launchvercel.com
Vercel launches Bring Your Own Cloud (BYOC) on AWS as GA
Vercel announced general availability of Bring Your Own Cloud (BYOC) on AWS, allowing Enterprise customers to run Vercel Functions within their own AWS accounts while Vercel manages deployments, routi…
- vercel.com
Vercel Connect adds Enterprise-Managed Authorization in public beta
Vercel Connect now offers Enterprise-Managed Authorization (EMA) in public beta, enabling apps and agents to access supported services on behalf of users without separate consent prompts. Admins manag…
- Featurevercel.com
Vercel adds TypeSafe AI's Jev 1.14 with image input and long-context support
Vercel now offers TypeSafe AI's Jev 1.14 on its AI Gateway, enabling image input for decision models and an experimental long-context mode for processing large documents or logs. The update expands AI…
- Contentvercel.com
Vercel explains short-term vs. long-term AI agent memory with Drives and Sandbox
Vercel published a technical guide distinguishing short-term and long-term memory for AI agents, emphasizing persistent storage via Vercel Drives and Sandbox for reusable project facts and decisions. …
- Contentvercel.com
Vercel explains five ways to add long-term memory to AI agents
Vercel published a technical guide outlining five methods to implement long-term memory in AI agents, including structured profiles, conversation summaries, task journals, searchable records, and pers…
- Launchvercel.com
Vercel launches Token Pass for Pro teams with $80 monthly AI model credit
Vercel introduced Token Pass in early access for Pro teams, offering $80 monthly credit for AI models like GLM 5.3 and Kimi K3 at $30 per month. The service integrates with AI Gateway, supports Zero D…
- Launchvercel.com
Vercel launches Python as a first-class runtime with general availability
Vercel announced Python as a first-class language and runtime with general availability, enabling developers to build APIs, full-stack apps, and AI agents using native Python SDKs. The platform now su…
- Featurevercel.com
Vercel Connect adds MCP Events support in public beta
Vercel Connect now supports MCP Events in public beta, enabling apps and agents to subscribe to webhook events from compatible MCP servers for tasks like summarizing messages or reviewing issues. The …
- Featurevercel.com
Vercel Daemons enters public beta for long-running workloads
Vercel launched Vercel Daemons in public beta, enabling long-running processes (e.g., Celery workers, Temporal, CI runners) to run as containers on Fluid compute alongside apps. Users configure replic…
- vercel.com
Vercel introduces durable AI agents with Workflow SDK and Sandbox
Vercel launched durable AI agents that preserve execution progress across failures and long waits, enabling recovery from crashes and human approvals after initial requests end. The feature integrates…
- Featurevercel.com
Vercel introduces persistent workspaces for AI agents with Sandbox and Drives
Vercel detailed how its persistent workspace features—Vercel Drives and persistent Sandboxes—enable AI agents to retain editable files across sessions for tasks like coding, data analysis, research, c…
- Integrationvercel.com
Build a better storefront: Introducing Next.js Commerce for SFCC
Vercel introduced a new B2C Next.js Commerce template for Salesforce Commerce Cloud (SFCC), enabling teams to build high-performing, customizable, and SEO-friendly storefronts by combining SFCC's back…
- Featurevercel.com
Vercel hosts v0 workshop to build and deploy AI chatbot with Figma integration
Vercel introduced a hands-on workshop demonstrating how to use v0 to generate full-stack applications from Figma designs, integrate AI providers via the AI SDK, and deploy directly to Vercel. The sess…
- Contentvercel.com
Vercel explains when to use Queues vs. Workflows in backend architecture
Vercel published a guide distinguishing message queues (for independent, idempotent tasks) from durable workflows (for multi-step, stateful processes) and how each maps to Vercel Queues and Vercel Wor…
- Featurevercel.com
Vercel enables Python AI agents in Next.js with FastAPI and AI SDK UI
Vercel published a guide showing how to connect a Python AI agent to a Next.js app using FastAPI and AI SDK UI, enabling streaming responses and tool execution. The setup leverages Vercel's AI Gateway…
- Contentvercel.com
Vercel publishes guide on hosting Python APIs with deployment considerations
Vercel’s official guide compares hosting options for Python APIs by framework support, execution limits, and storage needs, highlighting Vercel Functions for supported Python web apps. It emphasizes t…
- vercel.com
Vercel launches Token Pass for Pro teams: $80 monthly credit on featured open-weight models
Vercel introduced Token Pass, a $30-per-team monthly subscription granting $80 in credits for featured open-weight models via AI Gateway. The feature includes zero data retention (ZDR) and supports mo…
- Securityvercel.com
Vercel Sandbox challenge uncovers Linux kernel zero day and $242K in bounties
Vercel’s $1M sandbox security challenge revealed a critical Linux kernel zero day allowing host memory leaks, crashes, and code execution, alongside a cross-tenant disk snapshot restore flaw. No custo…
- vercel.com
Vercel adds free Nemotron 3.5 Lightning to AI Gateway
Vercel now offers NVIDIA’s Nemotron 3.5 Lightning model for free on its AI Gateway as `nvidia/nemotron-3.5-lightning-free`, targeting long-running agents, chatbots, and RAG systems. The 30B-parameter …
- Contentvercel.com
Vercel explains 6 backend workload types and where each runs best
Vercel published a guide categorizing backend workloads into six types—interactive APIs, independent jobs, durable workflows, scheduled jobs, realtime connections, and compute-intensive tasks—and mapp…
- Contentvercel.com
Vercel highlights Python backend use cases for Next.js apps
Vercel published a guide outlining seven scenarios where a Python backend enhances Next.js applications, including AI assistants, data analysis, document processing, and geospatial features. The post …
- vercel.com
Vercel publishes guide on human review workflows for enterprise AI decisions
Vercel released a technical guide outlining how to design human review processes for enterprise AI decisions, emphasizing clear reviewer actions, evidence-based workflows, and durable state management…
- Featurevercel.com
Vercel speeds up repository checkout and Git LFS downloads by up to 54%
Vercel reduced repository checkout and Git LFS download times by 20% on average and 54% for LFS-enabled projects, cutting ~13 seconds per deployment. The improvements apply automatically to all deploy…
- Launchvercel.com
Vercel adds Microsoft MAI-Image-2.6 and MAI-Image-2.6-Flash to AI Gateway
Vercel now offers Microsoft’s MAI-Image-2.6 and MAI-Image-2.6-Flash on its AI Gateway, enabling text-to-image generation and image editing via the AI SDK, AI CLI, or OpenAI-compatible API. Both models…
- vercel.com
Vercel AI Gateway adds Google's Gemini 4 Argon model
Vercel integrated Google's Gemini 4 Argon model into its AI Gateway, enabling developers to call the model via Vercel's unified API. The model supports a 1M token context window, multi-modal inputs, a…
- Contentvercel.com
Vercel publishes guide on governing employee-built AI apps and agents
Vercel released a detailed framework for governing employee-built AI apps and agents, emphasizing ownership, access rules, data controls, spending limits, and pre-deployment reviews. The guide aligns …
- Contentvercel.com
Vercel introduces decision models for enterprise apps with AI Gateway
Vercel published a guide on using decision models in enterprise apps to classify requests (e.g., software, equipment, services) via AI, with human review and an AI SDK example. The post highlights sep…
- Contentvercel.com
Vercel publishes guide on testing AI decisions before automating enterprise workflows
Vercel released a comprehensive guide outlining how enterprises can test AI-driven decisions in workflows before full automation. The post details labeled case testing, parallel pilots, error metrics,…
- Launchvercel.com
Vercel adds Drama 3 text-to-speech model to AI Gateway with 80% discount
Vercel launched Drama 3, a text-to-speech model from Fish Audio, on its AI Gateway platform with an 80% discount for 30 days. The model supports natural-language voice direction, multi-speaker dialogu…
- Contentvercel.com
Vercel explains citizen development and promotes Vercel for Enterprise Apps and Agents
Vercel published a guide defining citizen developers as non-IT employees who build apps using low-code tools and AI coding assistants. The post highlights how Vercel for Enterprise Apps and Agents ena…
- Contentvercel.com
Vercel outlines seven internal AI agent examples for enterprise teams
Vercel published a guide detailing seven hypothetical internal AI agent use cases—such as answering company questions in Slack, preparing sales updates, and analyzing datasets—designed to help teams p…
- Contentvercel.com
Vercel publishes guide on auditing AI agents with OpenTelemetry spans
Vercel released a technical guide outlining how to build AI agent audit trails by connecting identity, tool calls, approvals, and results to reconstruct actions and investigate mistakes. The post emph…
- vercel.com
Vercel publishes guide on managing shadow AI and employee-built agents
Vercel released a comprehensive guide defining shadow AI as unapproved AI use at work, including employee-built agents, and outlining steps to identify, assess, and govern such tools. The post emphasi…
- Integrationvercel.com
Vercel adds Meta's Muse Wonder to AI Gateway for coding agents
Vercel integrated Meta's Muse Wonder model into its AI Gateway, enabling developers to access the 1M-token context model for coding, multi-step agent tasks, and computer use via a single API key. The …
- vercel.com
Vercel adds Microsoft Decision-1 to AI Gateway for structured decision-making
Vercel integrated Microsoft Decision-1 into its AI Gateway platform on October 9, 2026, enabling developers to use the model for structured decision tasks like categorizing feedback or assessing workf…
- Featurevercel.com
Vercel CLI adds domain purchase automation for agents
Vercel’s CLI now lets agents discover, check availability, price, and initiate domain purchases via new commands like `vercel domains search`, `vercel domains check`, `vercel domains price`, and `verc…
- Integrationvercel.com
Vercel AI Gateway adds Liquid AI's d1 decision model with vision support
Vercel integrated Liquid AI's d1 decision model into AI Gateway, enabling structured classification, routing, and scoring without text generation. The model supports both text and image inputs, access…
- Pricingvercel.com
Vercel Pro teams default to 30-day deployment retention, storage costs unchanged
Vercel’s Pro teams now default to a 30-day deployment retention period to manage storage costs amid increased deployment frequency. Users can extend retention in settings, while storage remains billed…
- Case Studyvercel.com
Rillet ships 3× faster with AI agents on Vercel
Rillet, an AI-native ERP platform, tripled its shipping rate in three months by deploying AI agents on Vercel using eve, Vercel Connect, and AI Gateway. A two-person team closed 800+ Linear tickets in…
- Pricingvercel.com
Vercel introduces Deployment Storage billing for Pro teams at $0.10/GB-month
Vercel will begin billing existing Pro teams for Deployment Storage and Functions Storage at $0.10 per GB-month, with billing start dates communicated via email. The deployment retention window is als…
- vercel.com
Vercel adds Microsoft Decision-1 to AI Gateway for decision-making workflows
Vercel integrated Microsoft’s Decision-1 model into its AI Gateway, enabling applications to classify text, select workflow steps, or evaluate AI responses with probabilistic outputs. The model suppor…
- Integrationvercel.com
Vercel adds xAI's Grok Imagine Video 1.5 Lite to AI Gateway
Vercel integrated xAI's Grok Imagine Video 1.5 Lite into its AI Gateway, enabling text-to-video generation with native audio in resolutions up to 1080p. Developers can access the model via Vercel's AI…
- Featurevercel.com
Vercel Sandbox creation waits until ready before resolving
Vercel Sandbox's `Sandbox.create()` now resolves only after snapshot restoration completes, ensuring the sandbox is ready for commands or file operations immediately. Startup errors now reject the cre…
- Featurevercel.com
Vercel adds skipMiddlewareRequestBody to reduce Fast Origin Transfer usage
Vercel introduced a new `skipMiddlewareRequestBody` option in vercel.json/vercel.ts to skip sending request bodies to Routing Middleware, reducing Fast Origin Transfer usage and potentially improving …
- Featurevercel.com
Vercel AI Gateway adds OpenAI Ultrafast service tier for GPT-6 models
Vercel’s AI Gateway now supports OpenAI’s Ultrafast service tier for GPT 6 Astra and GPT 6.1 Sol, enabling faster output for interactive apps and coding workflows. Ultrafast is available via AI SDK, C…
- Integrationvercel.com
Vercel AI Gateway adds Step 5 Preview for agentic coding and financial analysis
Vercel AI Gateway now supports Step 5 Preview, a StepFun model optimized for agentic coding, professional knowledge work, and financial analysis. The model offers 1M-token context windows for text and…
- vercel.com
Vercel expands AI Gateway Decision API with five new use cases and models
Vercel introduced five practical tools built with AI Gateway's Decision API, leveraging models like GPT-6 Luna Decisions, Jev, Liquid d1, and Laya for tasks such as interpreting cancellation requests,…
- vercel.com
Vercel adds FLUX 3 Image to AI Gateway with editing and multi-resolution support
Vercel now offers FLUX 3 Image from Black Forest Labs on its AI Gateway, enabling text-to-image generation and image editing without requiring a separate Black Forest Labs account. The model supports …
- Contentvercel.com
Vercel showcases 7 OpenAI Decisions API use cases with AI Gateway
Vercel published a guide detailing seven practical use cases for OpenAI's Decisions API, including roadmap theme matching, incident handoff reviews, and documentation contradiction detection. The post…
- vercel.com
Vercel compares OpenAI and Perplexity Decisions APIs for developers
Vercel published a detailed technical comparison of OpenAI’s and Perplexity’s Decisions APIs, highlighting differences in request formats, answer types, and deployment options. The guide includes code…
- Featurevercel.com
Vercel updates AI SDK and AI Gateway with new decision-making APIs
Vercel introduced `experimental_decide` and `decisionModel()` in its AI SDK to replace deprecated `experimental_evaluate` and `evaluationModel()` functions. The update aligns with OpenAI's native Deci…
- vercel.com
Vercel AI Gateway adds confidence-based decision fallbacks in beta
Vercel introduced confidence-based decision fallbacks in AI Gateway, allowing applications to escalate uncertain AI decisions when confidence falls below a set threshold. The feature supports Choice, …
- Contentvercel.com
Vercel demonstrates GitHub PR triage with OpenAI's Decisions API
Vercel published a guide and Node.js example showing how to use OpenAI's Decisions API to automatically triage GitHub pull requests by review needs, such as identifying compatibility changes or migrat…
- Featurevercel.com
Vercel adds Claude Haiku 5.5 to AI Gateway with adaptive effort levels
Vercel’s AI Gateway now supports Anthropic’s Claude Haiku 5.5, optimized for high-volume, cost-sensitive tasks like summaries and database queries. The model introduces adaptive thinking with five eff…
- Featurevercel.com
Vercel adds Glyph Cluster reasoning model to AI Gateway in stealth mode
Vercel made the Glyph Cluster reasoning model available for free during its stealth period on AI Gateway for Pro and Enterprise plan teams with AI Gateway credits. The model supports multi-step coding…
- vercel.com
Vercel adds OpenAI Decisions API to AI Gateway
Vercel integrated OpenAI's Decisions API into its AI Gateway, enabling users to route, triage, and score inputs via a `/v1/decisions` endpoint. The feature supports predicate, choice, and score questi…
- Featurevercel.com
Vercel makes Microfrontends Routing free for WAF-mitigated traffic
Vercel now offers Microfrontends Routing at no cost for traffic handled by its Web Application Firewall (WAF), including denied, challenged, or rate-limited requests. CDN Requests and Fast Data Transf…
- Featurevercel.com
Vercel adds timestamp attributes to Vercel Flags for time-based targeting
Vercel Flags now supports timestamp attributes for entities, enabling time-based targeting rules like date-range promotions or user registration cutoffs. Developers can configure these attributes via …
- Featurevercel.com
Vercel adds AI-powered CSV column mapping with Jev and Luna fallbacks
Vercel introduced a new feature enabling AI-driven mapping of CSV column headers to application fields using AI Gateway decision fallbacks. The system suggests mappings via Jev, escalates to GPT-6 Lun…
- Featurevercel.com
Vercel expands Jev use cases for AI applications with decision-making workflows
Vercel published nine practical use cases for Jev, its TypeSafe AI System One model, demonstrating how it can route forms, prioritize tickets, review tool calls, select models, categorize documents, a…
- vercel.com
Vercel adds AI Gateway decision fallbacks for translation review workflows
Vercel introduced a new AI Gateway feature that uses decision fallbacks to flag UI copy changes requiring translation review. The system compares old and new source text to detect meaning changes (e.g…
- vercel.com
Vercel AI Gateway adds Mistral Large 4 with multimodal support
Vercel AI Gateway now supports Mistral Large 4, an open-weight multimodal model combining text and image understanding, accessible via a single API key without requiring a separate Mistral account. Th…
- Featurevercel.com
Vercel adds Nano Banana 2.1 to AI Gateway with improved image editing
Vercel now supports Google’s Nano Banana 2.1 on AI Gateway, enhancing image generation and editing with product recontextualization, mask/ink-based edits, and improved factuality. The model delivers p…
- Featurevercel.com
Vercel’s AI Gateway now supports confidence-based decision fallbacks, allowing requests to escalate to a secondary model when the primary model’s confidence drops below a set threshold. Conditions can
- vercel.com
Vercel compares TypeSafe AI's Jev and Perplexity's Decisions API for AI workflows
Vercel published a detailed technical comparison of TypeSafe AI's Jev and Perplexity's Decisions API, focusing on input support, question types, deployment options, and migration considerations. The g…
- Featurevercel.com
Vercel compares Jev, Laya, and Liquid d1 decision models for AI Gateway
Vercel published a detailed comparison of three decision models—Jev, Laya, and Liquid d1—available via AI Gateway and direct integrations. Laya uniquely offers self-hosting with downloadable checkpoin…
- Integrationvercel.com
Vercel publishes guide to integrating Perplexity's Decisions API
Vercel published a technical guide explaining how to integrate Perplexity’s Decisions API into applications, covering typed predictions, image support, and workflow design. The API evaluates text and …
- Integrationvercel.com
Vercel demonstrates Perplexity's Decisions API integration for image classification
Vercel published a technical guide showing how to use Perplexity's Decisions API to classify support screenshots by routing them to specialist queues based on image content and customer messages. The …
- Featurevercel.com
Vercel adds experimental Jev classifier API to Python AI SDK
Vercel introduced an experimental `evaluate()` API in its Python AI SDK to interact with the Jev model, a universal classifier designed for structured decision-making. The update enables Python develo…
- Case Studyvercel.com
Rogo ships 73K deployments/month with agent-written code in 5 minutes on Vercel
Rogo, an AI agent platform for financial institutions, deployed 73,000 times in a single month on Vercel, with agent-written code shipping to production in just 5 minutes. Their production AI agents u…
- Integrationvercel.com
Vercel integrates OpenAI DevDay 2026 features into v0, AI Gateway, Codex Cloud, and Agents API
Vercel announced integration of OpenAI DevDay 2026 features into its ecosystem, including GPT-6.1 Sol via AI Gateway, Codex Cloud for remote dev tasks, Agents API for browser automation, and v0 suppor…
- Contentvercel.com
Vercel explains decision models and their role in AI workflows
Vercel published a guide defining decision models as AI systems that evaluate context against structured questions to return typed judgments (Choice, Score, Boolean) with probabilities. The post highl…
- Featurevercel.com
Vercel adds Laya decision model to AI Gateway with free access through October 31, 2026
Vercel introduced the Laya decision/evaluation model from Convai Innovations on its AI Gateway, offering free access via Boundless through October 31, 2026. The model supports structured decision-maki…
- Integrationvercel.com
Vercel AI Gateway adds Microsoft AI models for speech and transcription
Vercel integrated Microsoft AI's latest audio models (MAI-Voice-2.1, MAI-Voice-2.1-Flash, and MAI-Transcribe-2 Streaming) into its AI Gateway platform. The models enable expressive multilingual speech…
- Sunsetvercel.com
Vercel Speed Insights deprecates First Input Delay on November 1st
Vercel's Speed Insights will stop recording First Input Delay (FID) measurements on November 1, 2026, as it transitions to Interaction to Next Paint (INP) as the responsiveness metric. No action is re…
- Integrationvercel.com
Vercel enables ChatGPT plugin extensions with MCP Apps and Next.js starter
Vercel published guidance on building ChatGPT plugin extensions using MCP Apps, including sidebar apps, conversation panels, and file viewers. The company provides a Next.js starter template and hosti…
Track Vercel on autopilot
- · Weekly AI brief: narrative summary of what shipped, every Monday 9 AM
- · Email or Slack alerts, or chat with the archive in your dashboard
- · Add Vercel + up to 2 more competitors free, no credit card