---
name: Vercel
slug: vercel
type: scrape
source_url: https://vercel.com/changelog
organization: Vercel
organization_slug: vercel
total_releases: 54
latest_date: 2026-06-23
last_updated: 2026-06-23
tracking_since: 2026-03-18
canonical: https://releases.sh/vercel/vercel
organization_url: https://releases.sh/vercel
---

<Summary type="rolling" window-days="90" release-count="14">
Vercel's platform moved toward tighter control and flexibility across execution, caching, and feature management. The company graduated Vercel Flags to general availability, adding framework-native flag definitions for Next.js and SvelteKit while supporting the OpenFeature standard for other stacks. The AI Gateway expanded significantly—Claude Opus 4.7 arrived with task budgets and improved agentic execution, and Seedance 2.0 brought multimodal video generation (text-to-video, image-to-video, video editing) without requiring separate provider accounts. On the infrastructure side, new projects now respect upstream cache-control headers by default, Sandbox persistence moved to beta for automatic filesystem snapshots, and elastic build machines entered beta to right-size compute per project.
</Summary>

<Summary type="monthly" period="March 2026" release-count="11">
Vercel shipped infrastructure improvements and security responses across caching, build machines, and sandbox persistence. The Axios supply chain attack on March 31st prompted immediate remediation—blocking malicious versions and C&C access from build infrastructure—while new projects defaulted to respecting upstream cache headers starting April 6th. The SDK layer expanded with Sandbox automatic persistence (saving filesystem state on stop and restoring on resume), Chat SDK concurrent message handling and Slack scheduling, file permission control in sandbox writes, and Activity Log integration into the CLI, while elastic build machine assignment moved to beta to right-size compute per project.
</Summary>

<Release date="June 23, 2026" published="2026-06-23T00:00:00.000Z" url="https://vercel.com/changelog#deploy-from-claude-design-to-vercel">
## Deploy from Claude Design to Vercel

Deploy from Claude Design to Vercel
</Release>

<Release date="June 22, 2026" published="2026-06-22T00:00:00.000Z" url="https://vercel.com/changelog#websocket-support-is-now-in-public-beta">
## WebSocket support is now in Public Beta

WebSocket support is now in Public Beta
</Release>

<Release date="June 22, 2026" published="2026-06-22T00:00:00.000Z" url="https://vercel.com/changelog#vercel-cli-now-supports-signing-blob-urls">
## Vercel CLI now supports signing blob URLs

Vercel CLI now supports signing blob URLs
</Release>

<Release date="June 22, 2026" published="2026-06-22T00:00:00.000Z" url="https://vercel.com/changelog#increased-limit-for-projects-per-git-repo">
## Increased limit for projects per Git repo

Increased limit for projects per Git repo
</Release>

<Release date="June 22, 2026" published="2026-06-22T00:00:00.000Z" url="https://vercel.com/changelog#sakana-fugu-ultra-now-available-on-ai-gateway">
## Sakana Fugu Ultra now available on AI Gateway

Sakana Fugu Ultra now available on AI Gateway
</Release>

<Release date="June 17, 2026" published="2026-06-17T00:00:00.000Z" url="https://vercel.com/changelog#vercel-passport-is-now-in-public-beta">
## Vercel Passport is now in Public Beta

Vercel Passport is now available in Public Beta.
</Release>

<Release date="June 17, 2026" published="2026-06-17T00:00:00.000Z" url="https://vercel.com/changelog#introducing-eve-an-open-source-agent-framework">
## Introducing eve, an open-source agent framework

Eve is a new open-source agent framework from Vercel.
</Release>

<Release date="June 17, 2026" published="2026-06-17T00:00:00.000Z" url="https://vercel.com/changelog#cli-deployment-limits-removed">
## CLI deployment limits removed

CLI deployment limits have been removed.
</Release>

<Release date="June 17, 2026" published="2026-06-17T00:00:00.000Z" url="https://vercel.com/changelog#vercel-connect-secure-access-to-external-services-for-your-a">
## Vercel Connect: Secure access to external services for your agents

Vercel Connect enables secure access to external services for your agents.
</Release>

<Release date="June 16, 2026" published="2026-06-16T00:00:00.000Z" url="https://vercel.com/changelog#glm-5-2-now-available-on-ai-gateway">
## GLM 5.2 now available on AI Gateway

GLM 5.2 is now available on AI Gateway. Built for long-horizon tasks, GLM 5.2 carries project-level engineering context across a single task, runs long-running tasks more reliably, and follows engineering standards more consistently. The context window has been upgraded to 1M tokens, up from 200K in GLM 5.1. Available via the AI SDK with model ID `zai/glm-5.2`. AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations.
</Release>

<Release date="June 16, 2026" published="2026-06-16T00:00:00.000Z" url="https://vercel.com/changelog#workflow-sdk-now-supports-tanstack-start">
## Workflow SDK now supports TanStack Start

Workflow SDK now supports TanStack Start applications on Vercel. TanStack Start is built on Vite and Nitro, so the existing `workflow/vite` plugin works directly. Add it to `vite.config.ts` alongside `tanstackStart()`.
</Release>

<Release date="June 16, 2026" published="2026-06-16T00:00:00.000Z" url="https://vercel.com/changelog#vercel-sandbox-can-now-run-for-up-to-24-hours">
## Vercel Sandbox can now run for up to 24 hours

Vercel Sandboxes can run uninterrupted sessions for up to 24 hours (up from 5 hours). This unlocks workloads requiring longer runtimes, such as large-scale data processing, end-to-end testing pipelines, and long-lived agentic workflows. The 24 hour max duration is available on all Pro and Enterprise plans. Can be paired with persistent sandboxes to maintain durable state across extended runs.
</Release>

<Release date="June 16, 2026" published="2026-06-16T00:00:00.000Z" url="https://vercel.com/changelog#workflow-sdk-now-supports-inflight-cancellation">
## Workflow SDK now supports inflight cancellation

The Workflow SDK 5 beta now supports the standard `AbortController` and `AbortSignal` APIs across workflow and step boundaries. Create a controller inside a workflow, pass its signal into steps, and cancel in-flight operations using the same API `fetch` already uses. The signal stays durable across suspensions and deterministic replay. Cancellation is cooperative; steps have to inspect the signal or pass it to an API that supports `AbortSignal`. Use cases include stopping slow steps on timeout, cancelling remaining requests after first success, threading signals through pipelines, or cancelling parallel work on external condition changes.
</Release>

<Release date="June 15, 2026" published="2026-06-15T00:00:00.000Z" url="https://vercel.com/changelog#increased-blob-store-limit-for-hobby-users">
## Increased Blob store limit for Hobby users

Hobby users can now create up to 100 Blob stores, up from 5. This gives teams more flexibility to organize data by project, environment, or region as applications grow. Storage, operations, and transfer limits still apply.

Learn more in the [Blob documentation](https://vercel.com/docs/vercel-blob/usage-and-pricing).
</Release>

<Release date="June 13, 2026" published="2026-06-13T00:00:00.000Z" url="https://vercel.com/changelog#workflow-sdk-now-runs-natively-in-nitro-v3">
## Workflow SDK now runs natively in Nitro v3

Workflow SDK's native Nitro v3 integration is now in beta. Steps run inside the same bundled runtime as the rest of your app, instead of a separate bundle. Nitro's `useStorage()` and other server-side APIs work directly inside `"use step"` functions.

The Nitro dev server also serves the workflow web UI at `/_workflow`. Open it in your browser to inspect, monitor, and debug workflow runs.

Workflow routes are now bundled by Nitro as part of the app build. Dependencies are traced, and unused code is tree-shaken, so the output includes only what runs, with faster builds and smaller bundles.
</Release>

<Release date="June 12, 2026" published="2026-06-12T00:00:00.000Z" url="https://vercel.com/changelog#claude-fable-5-access-suspended-on-ai-gateway">
## Claude Fable 5 access suspended on AI Gateway

In compliance with the US Government's directive, access to Claude Fable 5 is now suspended on AI Gateway for all users. We do not know if or when access to the model will be restored.

You can still access and use other Anthropic models through AI Gateway.
</Release>

<Release date="June 12, 2026" published="2026-06-12T00:00:00.000Z" url="https://vercel.com/changelog#kimi-k2-7-code-now-available-on-ai-gateway">
## Kimi K2.7 Code now available on AI Gateway

Kimi K2.7 Code from Moonshot AI is now available on AI Gateway. K2.7 Code is a coding model built for long-horizon programming tasks, generalizing across scenarios including frontend development, DevOps, and performance optimization. The model has a native multimodal architecture that supports text and vision input, and always runs in thinking mode.

To use K2.7 Code, set model to `moonshotai/kimi-k2.7-code` in the AI SDK. The model supports both text and image inputs for multimodal capabilities.
</Release>

<Release date="June 12, 2026" published="2026-06-12T00:00:00.000Z" url="https://vercel.com/changelog#program-claude-code-codex-pi-and-other-agent-harnesses-with-">
## Program Claude Code, Codex, Pi and other agent harnesses with AI SDK

AI SDK 7 introduces `HarnessAgent`, a single API for running established agent harnesses, including Claude Code, Codex, and Pi. Write the agent once and use the best harness available. Harnesses manage components including skills, sandboxes, sessions, permission flows, compaction, runtime configuration, and sub-agents. The AI SDK normalizes access to these capabilities through a unified harness abstraction. Initial harness adapters for this experimental release include Claude Code, Codex, and Pi, with more coming soon.
</Release>

<Release date="June 9, 2026" published="2026-06-09T00:00:00.000Z" url="https://vercel.com/changelog#budgets-for-api-keys-on-ai-gateway">
## Budgets for API keys on AI Gateway

Set spend caps on API keys for AI Gateway to control AI costs. Once a budget limit is exceeded, AI Gateway rejects further requests on that key until the budget resets or is raised. The cap applies to all AI Gateway providers and models running through the key.

Create budgeted API keys in the Vercel Dashboard via the AI Gateway API Keys page, or programmatically through the Vercel CLI:

```
vercel ai-gateway api-keys create --name <NAME> --budget <DOLLARS> --refresh-period <PERIOD>
```

Refresh periods can be `daily`, `weekly`, `monthly`, or `none`. Each period resets at the start of its window in UTC.
</Release>

<Release date="June 9, 2026" published="2026-06-09T00:00:00.000Z" url="https://vercel.com/changelog#claude-fable-5-now-available-on-ai-gateway">
## Claude Fable 5 now available on AI Gateway

Claude Fable 5 from Anthropic is now available on AI Gateway. A Mythos-class model, Fable 5 is a notable step up over prior Claude models on long-running, ambiguous, multi-step tasks. The model sustains productive output across multi-day runs and dependably dispatches parallel sub-agents. Code review, bug-finding, and repository investigation are stronger, and first-shot correctness on complex problems is noticeably higher.

Fable 5 ships with blocking classifiers that refuse offensive cybersecurity, biology, and summarized-thinking extraction. Anthropic does not support Zero Data Retention; prompts and completions are retained for 30 days and are not used to train Claude.

To use Fable 5, set the model to `anthropic/claude-fable-5` in the AI SDK.
</Release>

<Pagination cursor="2026-06-09T00:00:00.000Z|2026-06-10T01:02:13.022Z|rel_r9ejeZ4TeWa9FwClr-dnM" next="https://releases.sh/vercel/vercel.md?cursor=2026-06-09T00%3A00%3A00.000Z%7C2026-06-10T01%3A02%3A13.022Z%7Crel_r9ejeZ4TeWa9FwClr-dnM&limit=20" />