AI Gateway has a unified fast mode abstraction, now in beta. You can now request fast mode the same way for every model on AI Gateway. Set speed to fast , and the gateway serves the fast tier when it's available and falls back to standard speed when it isn't. Fast mode trades a higher per-token cost for lower latency or higher throughput.
Vercel Passport is now generally available. Passport allows you to protect your Vercel deployments with your own identity provider. Visitors authenticate through Okta, Microsoft Entra ID, or any OIDC provider before viewing a protected deployment, and Vercel forwards a signed identity token to the deployment so application code can build on who the visitor...
On AI Gateway , GPT-5.6 Luna and GPT-5.6 Terra are now cheaper and GPT-5.6 Sol is faster. 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.
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.
Patch Changes dcadcad: Trigger a new release to publish versions that were skipped by the failed release run.
Pro and Enterprise teams can now purchase additional custom environment capacity without contacting sales. Custom environments let you model your team's release process on Vercel. You can add staging , qa , or any named stage between preview and production , and each environment gets its own branch tracking, environment variables, and domains.
Grok Voice Think Fast 2.0 from xAI is now available on AI Gateway. It is a speech-to-speech voice model that takes audio in and audio out, improving on the previous Grok Voice model in reasoning, transcription accuracy, and conversation. The model reasons in parallel with speech, so it can think through a query while talking without adding latency.
Minor Changes 0f04613: Add vercel vcr permissions subcommands for managing which teams can pull images from a VCR repository: vcr permissions ls lists teams with access (by team slug), add and rm grant or revoke access for one or more teams (comma-separated team ids or slugs), and clear removes access for every team.
Vercel Connect now lets you link a connector to a Custom Environment so deployments there can request provider tokens and receive forwarded webhooks. Custom Environments appear alongside Production, Preview, and Development when you add or edit a project on a connector in the dashboard.
Vercel Sandbox now supports forking with Sandbox.fork() . The fork starts from the source's current snapshot and inherits its config and environment variables. If the source is running, it forks the latest saved state, not the live in-memory state. If the source has no snapshot, it falls back to a fresh create, using the source's runtime and config.
Minor Changes be40159: Extend vercel ai-gateway api-keys create with --expiration and --alert-thresholds flags for parity with the create endpoint. 4631896: Add vercel ai-gateway api-keys list, inspect, and remove subcommands so API keys can be listed, inspected, and deleted from the CLI.
Kimi K3 from Moonshot AI and its faster serving path, Kimi K3 Fast , are now available from US-based providers on AI Gateway, including Baseten and Fireworks. Zero Data Retention (ZDR) is also supported for both models. Running Kimi K3 on US-based providers lets teams with data residency and compliance requirements use the model on US infrastructure.
The Nuxt team has released Nuxt 4.5.1 and 3.21.10, along with @nuxt/devtools 3.3.1, to address eight security advisories, including a high-severity server-side remote code execution vulnerability.
Minor Changes 843bb99: Support attaching Connect connectors to project custom environments by slug or stable ID. d277bd9: Add custom-environment targeting for Connex trigger destinations. 4502520: Support Node.js Routing Middleware entrypoints through proxy.entrypoint, with optional path matching through proxy.matcher.
Workflow steps on Pro and Enterprise plans can now run for up to 30 minutes (1800 seconds), up from 800 seconds, using extended function durations (in beta). To opt in, set VERCEL_ENABLE_WORKFLOW_EXTENDED_MAX_DURATION to 1 in your project's Environment Variables, then redeploy. Requires Fluid compute and a supported Node.js or Python runtime.
The Vercel WAF can now protect a Vercel Blob store. The same rules that guard your deployments (deny, challenge, rate limit) now apply to blob traffic with no changes to your code, blob URLs, or @vercel/blob . Every blob is already served through Vercel's CDN , so protection is a switch on the store, not a new proxy.
Minor Changes 7dd4301: Make service bindings type optional. Patch Changes 3a066fe: Add a resolved attribute to the vc.importBuilders trace span recording each Builder's resolved version and source directory b40db5b: Improve vc alerts and vc alerts rules output for custom alerts, including readable inspect views, custom alert rule filtering, compact rule...
Ling 3.0 Flash from Ant Group is now available on AI Gateway. The model is free to use for the next three weeks, through August 3rd. Ling 3.0 Flash is a Mixture-of-Experts model with 124B total parameters and about 5.1B active per token. It has a 256K token context window and runs in thinking and non-thinking modes.
Vercel Flags now shows a live evaluation view on each flag's detail page. You can see evaluations per minute charted over time, with each flag version marked in the chart so you can tie evaluation shifts to specific configuration changes. You can group and filter evaluations by variant, reason, environment, SDK key, client, and reporting project.
AI Gateway now supports streaming transcription . Previously, transcription required a complete audio file and returned the full transcript in a single response. Now you can stream audio in as it's captured and get transcript updates back as the model produces them, keeping latency low for uses like live captioning and voice input.
Minor Changes ab9c775: Add vercel ai-gateway budgets set/list/remove commands Patch Changes 83c72a4: Recover from sensitive Environment Variable step-up challenges in every command that fetches Environment Variables (vercel dev, vercel pull, vercel env run, vercel build, vercel link, OIDC token refresh), not just vercel env pull.
Gemini 3.6 Flash and Gemini 3.5 Flash-Lite are now available on AI Gateway. Gemini 3.6 Flash improves quality across coding, agentic tasks, and web development while consuming fewer tokens and making fewer model calls. It produces cleaner web and app development output.
Vercel now compiles Python functions to bytecode at build time. In our benchmarks, cold starts for the median-sized function dropped from 2.8s to 1.3s . Cold starts are already rare in practice.
Vercel Workflows now keeps each run's state, queue dispatch, and output streams in a single home region: the region where the run starts by default, or any target region you choose.
Team Owners can now clear the team's Remote Cache of all artifacts in one click. This is useful when you believe there are poisoned artifacts in your cache. In your team's Build and Deployment settings, visit the Remote Caching section and clear the Remote Cache. Visit the docs to learn more. Read more
Minor Changes ed28e18: Add vercel flags evaluations for viewing compact, per-variant evaluation charts and bucket-level JSON data with truncation metadata. Examples: bash vercel flags evaluations my-feature vercel flags evaluations my-feature --since 24h --granularity 1h vercel flags evaluations my-feature --since 24h --json Patch Changes 37a95cd: Include...
Patch Changes 145eb34: Improve vercel ai-gateway coding-agents setup prompts: use the searchable team picker, keep the multiselect key legend visible after the first selection, and narrow the prompt lines by moving context off the question line.
Vercel now reuses static files across deployments for frameworks that output content-addressed assets. It's zero-config: Vercel uses Framework-defined infrastructure to manage these immutable files alongside your code changes, handling the hard parts like hash collisions, file lifecycles, and efficient routing.
GLM 5.2 is 35% off on AI Gateway through July 24 when routed through Novita. To get the discounted rate, set the model to zai/glm-5.2 in the AI SDK and route requests through Novita: After July 24, the model stays available at standard provider rates with no markup. Try GLM 5.2 in the model playground . Read more
Patch Changes b08b736: Support linked-project deploys with project-scoped tokens that cannot fetch owner user/team resources.
Kimi K3 from Moonshot AI is now available on AI Gateway. K3 is an open-source model with a 1M-token context window and native visual understanding, accepting text, image, and video inputs.
The ISR Observability page now shows Observability Plus subscribers a write utilization metric, helping identify routes that regenerate often but receive few requests. Write utilization is the ratio of cached requests to ISR writes.
Minor Changes 238543c: Support Python services that declare their web app and development workflow sidecars through a pyproject.toml entrypoint in vercel dev. Patch Changes d5751d6: Allow Env, Target, Flags rules, Routes, Redirects, Firewall, and Git commands to select a project explicitly with --project, including from an unlinked directory.
The trace viewer for Vercel Workflows now has a minimap that shows the entire run in one view. Drag the viewport to pan, resize it to zoom, or drag across the minimap to select a new section. Click anywhere on the minimap to jump to that point, or scroll and use arrow keys to step through the run.
You can now send Discord bot messages with Components V2 , an opt-in layout system that treats text, images, files, and buttons as flexible components you can arrange in any order. Set the adapter's contentFormat option to ComponentsV2 and cards render with native containers, sections, media galleries, separators, buttons, and string selects.
Patch Changes ce094f1: Show a specific error message when purchasing v0 credits without a paid v0 plan, instead of the generic unexpected-error fallback. 3e0e6a8: Refactor linked project resolution to use named options. 877c1ba: Allow vercel env ls to target a project explicitly without linking the current directory.
Endform is now available on the Vercel Marketplace . It runs your Playwright tests in parallel, so your suite finishes in the time of your slowest test. Key capabilities: Run every test on its own isolated machine Bring your existing Playwright tests with no config changes Get a pass/fail check on every production and preview deployment, linked to full...
We are making the data behind the AI Gateway leaderboards open under the CC BY 4.0 license . You can now download or query the data through the leaderboard-export API endpoint and render any chart as a shareable image. The AI Gateway leaderboards show how AI is used in production, ranking traffic for models, labs, apps, and providers.
Minor Changes 997d319: vercel env add now accepts a comma-separated list of environments to create a single Environment Variable entry spanning multiple targets, e.g. vercel env add API_URL production,preview,development.
You can now inspect subagent activity for eve projects in Agent Runs . The new Subagents tab shows every subagent, organized by which turn started it. Each row shows the prompt, duration, and any failures, all on a shared timeline. Click any subagent to open its run.
Deployment Policies are now available and enable teams to restrict which mechanisms, organizations, and repositories are allowed to create deployments. Each policy can be configured per environment at the team and project level, allowing flexible combinations of rules. Learn more about Deployment Policies . Read more
Major Changes 4998ec8: vercel routes export now uses --output (-o) instead of --format to select the output file format (json or ts, dotted forms like .ts also accepted). This frees --format to align with the standard output-format convention used by other commands. Update any scripts from routes export --format ts to routes export --output ts.
Seedream 5.0 Pro is now available on AI Gateway . Seedream 5.0 Pro is an image generation and editing model. It generates images from text, rendering text without spelling errors and following typographic rules, and produces dense infographics with charts, timelines, and layouts alongside realistic imagery.
Traces in Vercel Logs now support Tree and Waterfall views. See span hierarchy, critical path, and timing without leaving the log entry. The Tree view groups spans by parent and child. On load, spans are sorted by duration so the slowest appear first.
GPT 5.6 is now available on AI Gateway in three models: Sol, Terra, and Luna. GPT 5.6 from OpenAI is now available on AI Gateway in a limited preview, across three models: Sol, Terra, and Luna. All three are stronger at agentic work across coding, biology, and cybersecurity, and are more token-efficient than the previous generation.
When an Environment Variable is marked Sensitive and has a value 32 characters or longer, Vercel now replaces it with [REDACTED] in the deployment's build logs. The Build Logs view also indicates when redaction happened, giving teams visibility into why output was masked without ever exposing the underlying value.
Major Changes 8f10c0b: Require an explicit team signal when linking without a TTY. In non-interactive mode or without a terminal, vercel link (and other commands that set up a link) no longer fall back to the globally selected team from vc switch or the login default, and --yes no longer guesses a team.
Grok 4.5 from SpaceXAI is now available on AI Gateway . Built for coding, knowledge work, and STEM, the model accepts text and image inputs. Grok 4.5 supports low, medium, and high reasoning levels and defaults to high. Set the level with reasoning to balance speed against depth.
Chat SDK now supports Dial with the new vendor-official adapter . Build bots that send and receive SMS, MMS, and iMessage on a real phone number, with bidirectional media and inbound voice-call transcripts. Replies use the standard Chat SDK thread and message APIs, with HMAC-verified webhooks and stable per-conversation threading.
GitHub Tools now ships an eve toolset through the new @github-tools/sdk/eve subpath. One file in agent/tools/ can register every GitHub tool, or use a preset such as maintainer , so you can build a complete GitHub agent in nine lines of code. Safe by default: Every write tool, such as mergePullRequest , requires approval unless you opt out.
Project Settings now includes its own project-specific A ctivity Log . This view pulls the relevant user events from the Activity Log in Team Settings. Review project-level activity without leaving your project settings. Visit your project settings to try it out. Read more
Patch Changes 1e774e8: Preserve generated service routes when Build Output services are merged from framework output into a custom build output directory. c9c9edb: Add --project support to project-scoped flags, cache, crons, and rolling-release commands. @vercel/static-build@2.11.4
Minor Changes cc74129: Enable monorepo subdirectory build fixes by default (previously gated behind VERCEL_RESOLVE_ROOT_DIRECTORY=1), scoped to directories the workspace actually claims.
Your agent can now inspect Agent Runs via the Vercel MCP and CLI for eve , the open-source agent framework. eve traces are automatically ingested when deployed to Vercel and available as Agent Runs.
Vercel Flags segments can now be managed from the Vercel CLI with the new vercel flags segments command. A segment is the targeting primitive a flag uses to decide who sees what. Membership composes from three repeatable tokens: include: , exclude: , and rule: . Pass them to --add or --remove for incremental edits.
Minor Changes b9e4f06: Add a vercel agent-runs command for Agent Runs observability: list lists Agent Runs for a project, inspect shows run metadata, events, usage, and subagents, trace shows the full trace (turns, messages, reasoning, and tool calls), and projects lists projects in the team with Agent Runs activity.
Vercel AI Gateway now supports routing rules . Routing rules are firewall-style rules that control which models your team can use, applied at the gateway level instead of in your application code. When a model goes down or gets retired, you usually have to ship a code change to move off it.
Minor Changes 2c87793: Add vercel vcr commands to manage Vercel Container Registry repositories and images. Patch Changes 84c01ff: Add install as an alias for vercel integration add.
You can now preview the framework preset and files that Vercel CLI includes in a deployment before creating one. Run vercel deploy --dry from a linked project: For automation or further inspection, return the complete file manifest as JSON: JSON output includes the detected framework, included and ignored paths, directory size distribution, largest files...
Access to Claude Fable 5, the Mythos-class model, has now been restored on AI Gateway following the US Government's decision to lift the export controls. Fable 5 is the same model that was available between June 9 and June 12. What has changed is the safety classifiers, which are now updated and more robust.
Patch Changes 4d9394b: Bridge WebSocket close to response close c4667d3: Fix vc build run from a monorepo subdirectory (gated behind VERCEL_RESOLVE_ROOT_DIRECTORY=1). When a project is linked in place (apps/api/.vercel/project.json) and vc build is run from that directory, the build previously treated the linked subdirectory as the repository root.
Releasesby Glyph