Changelog

You can now create and manage Workflows using Terraform, now supported in the Cloudflare Terraform provider v5.11.0 ↗. Workflows allow you to build durable, multi-step applications -- without needing to worry about retrying failed tasks or managing infrastructure.

Now, you can deploy and manage Workflows through Terraform using the new cloudflare_workflow resource ↗:

resource "cloudflare_workflow" "my_workflow" {
  account_id    = var.account_id
  workflow_name = "my-workflow"
  class_name    = "MyWorkflow"
  script_name   = "my-worker"
}

Here are full examples of how to configure cloudflare_workflow in Terraform, using the existing cloudflare_workers_script resource ↗, and the beta cloudflare_worker_version resource ↗.

resource "cloudflare_workers_script" "workflow_worker" {
  account_id  = var.cloudflare_account_id
  script_name = "my-workflow-worker"

  content_file   = "${path.module}/../dist/worker/index.js"
  content_sha256 = filesha256("${path.module}/../dist/worker/index.js")
  main_module    = "index.js"
}

resource "cloudflare_workflow" "workflow" {
  account_id    = var.cloudflare_account_id
  workflow_name = "my-workflow"
  class_name    = "MyWorkflow"
  script_name   = cloudflare_workers_script.workflow_worker.script_name
}

You can more granularly control the lifecycle of each Worker resource using the beta cloudflare_worker_version ↗ resource, alongside the cloudflare_worker ↗ and cloudflare_workers_deployment ↗ resources.

resource "cloudflare_worker" "workflow_worker" {
  account_id = var.cloudflare_account_id
  name       = "my-workflow-worker"
}

resource "cloudflare_worker_version" "workflow_worker_version" {
  account_id = var.cloudflare_account_id
  worker_id  = cloudflare_worker.workflow_worker.id

  main_module         = "index.js"

  modules = [{
    name         = "index.js"
    content_file = "${path.module}/../dist/worker/index.js"
    content_type = "application/javascript+module"
  }]
}

resource "cloudflare_workers_deployment" "workflow_deployment" {
  account_id  = var.cloudflare_account_id
  script_name = cloudflare_worker.workflow_worker.name

  strategy = "percentage"
  versions = [{
    version_id = cloudflare_worker_version.workflow_worker_version.id
    percentage = 100
  }]
}

resource "cloudflare_workflow" "my_workflow" {
  account_id    = var.cloudflare_account_id
  workflow_name = "my-workflow"
  class_name    = "MyWorkflow"
  script_name   = cloudflare_worker.workflow_worker.name
}

• Get started with the Cloudflare Terraform provider (v5.11.0) ↗ and the new cloudflare_workflow resource ↗.

Each of your Workers now has a new overview page in the Cloudflare dashboard.

The goal is to make it easier to understand your Worker without digging through multiple tabs. Think of it as a new home base, a place to get a high-level overview on what's going on.

It's the first place you land when you open a Worker in the dashboard, and it gives you an immediate view of what’s going on. You can see requests, errors, and CPU time at a glance. You can view and add bindings, and see recent versions of your app, including who published them.

Navigation is also simpler, with visually distinct tabs at the top of the page. At the bottom right you'll find guided steps for what to do next that are based on the state of your Worker, such as adding a binding or connecting a custom domain.

We plan to add more here over time. Better insights, more controls, and ways to manage your Worker from one page.

If you have feedback or suggestions for the new Overview page or your Cloudflare Workers experience in general, we'd love to hear from you. Join the Cloudflare developer community on Discord ↗.

You can now enable compaction for individual Apache Iceberg ↗ tables in R2 Data Catalog, giving you fine-grained control over different workloads.

# Enable compaction for a specific table (no token required)
npx wrangler r2 bucket catalog compaction enable <BUCKET> <NAMESPACE> <TABLE> --target-size 256

This allows you to:

• Apply different target file sizes per table
• Disable compaction for specific tables
• Optimize based on table-specific access patterns

Learn more at Manage catalogs.

You can now enable Cloudflare Access for your workers.dev and Preview URLs in a single click.

Access allows you to limit access to your Workers to specific users or groups. You can limit access to yourself, your teammates, your organization, or anyone else you specify in your Access policy.

To enable Cloudflare Access:

1. In the Cloudflare dashboard, go to the Workers & Pages page.

   Go to Workers & Pages ↗

2. In Overview, select your Worker.

3. Go to Settings > Domains & Routes.

4. For workers.dev or Preview URLs, click Enable Cloudflare Access.

5. Optionally, to configure the Access application, click Manage Cloudflare Access. There, you can change the email addresses you want to authorize. View Access policies to learn about configuring alternate rules.

To fully secure your application, it is important that you validate the JWT that Cloudflare Access adds to the Cf-Access-Jwt-Assertion header on the incoming request.

The following code will validate the JWT using the jose NPM package ↗:

import { jwtVerify, createRemoteJWKSet } from "jose";

export default {
	async fetch(request, env, ctx) {
		// Verify the POLICY_AUD environment variable is set
		if (!env.POLICY_AUD) {
			return new Response("Missing required audience", {
				status: 403,
				headers: { "Content-Type": "text/plain" },
			});
		}

		// Get the JWT from the request headers
		const token = request.headers.get("cf-access-jwt-assertion");

		// Check if token exists
		if (!token) {
			return new Response("Missing required CF Access JWT", {
				status: 403,
				headers: { "Content-Type": "text/plain" },
			});
		}

		try {
			// Create JWKS from your team domain
			const JWKS = createRemoteJWKSet(
				new URL(`${env.TEAM_DOMAIN}/cdn-cgi/access/certs`),
			);

			// Verify the JWT
			const { payload } = await jwtVerify(token, JWKS, {
				issuer: env.TEAM_DOMAIN,
				audience: env.POLICY_AUD,
			});

			// Token is valid, proceed with your application logic
			return new Response(`Hello ${payload.email || "authenticated user"}!`, {
				headers: { "Content-Type": "text/plain" },
			});
		} catch (error) {
			// Token verification failed
			return new Response(`Invalid token: ${error.message}`, {
				status: 403,
				headers: { "Content-Type": "text/plain" },
			});
		}
	},
};

Add these environment variables to your Worker:

• POLICY_AUD: Your application's AUD tag
• TEAM_DOMAIN: https://<your-team-name>.cloudflareaccess.com

Both of these appear in the modal that appears when you enable Cloudflare Access.

You can set these variables by adding them to your Worker's Wrangler configuration file, or via the Cloudflare dashboard under Workers & Pages > your-worker > Settings > Environment Variables.

Deepgram's newest Flux model @cf/deepgram/flux is now available on Workers AI, hosted directly on Cloudflare's infrastructure. We're excited to be a launch partner with Deepgram and offer their new Speech Recognition model built specifically for enabling voice agents. Check out Deepgram's blog ↗ for more details on the release.

The Flux model can be used in conjunction with Deepgram's speech-to-text model @cf/deepgram/nova-3 and text-to-speech model @cf/deepgram/aura-1 to build end-to-end voice agents. Having Deepgram on Workers AI takes advantage of our edge GPU infrastructure, for ultra low latency voice AI applications.

For the month of October 2025, Deepgram's Flux model will be free to use on Workers AI. Official pricing will be announced soon and charged after the promotional pricing period ends on October 31, 2025. Check out the model page for pricing details in the future.

The new Flux model is WebSocket only as it requires live bi-directional streaming in order to recognize speech activity.

1. Create a worker that establishes a websocket connection with @cf/deepgram/flux

export default {
  async fetch(request, env, ctx): Promise<Response> {
    const resp = await env.AI.run("@cf/deepgram/flux", {
      encoding: "linear16",
      sample_rate: "16000"
    }, {
      websocket: true
    });
    return resp;
  },
} satisfies ExportedHandler<Env>;

2. Deploy your worker

npx wrangler deploy

3. Write a client script to connect to your worker and start sending random audio bytes to it

const ws = new WebSocket('wss://<your-worker-url.com>');

ws.onopen = () => {
  console.log('Connected to WebSocket');

  // Generate and send random audio bytes
  // You can replace this part with a function
  // that reads from your mic or other audio source
  const audioData = generateRandomAudio();
  ws.send(audioData);
  console.log('Audio data sent');
};

ws.onmessage = (event) => {
  // Transcription will be received here
  // Add your custom logic to parse the data
  console.log('Received:', event.data);
};

ws.onerror = (error) => {
  console.error('WebSocket error:', error);
};

ws.onclose = () => {
  console.log('WebSocket closed');
};

// Generate random audio data (1 second of noise at 44.1kHz, mono)
function generateRandomAudio() {
  const sampleRate = 44100;
  const duration = 1;
  const numSamples = sampleRate * duration;
  const buffer = new ArrayBuffer(numSamples * 2);
  const view = new Int16Array(buffer);

  for (let i = 0; i < numSamples; i++) {
    view[i] = Math.floor(Math.random() * 65536 - 32768);
  }

  return buffer;
}

You can now perform more powerful queries directly in Workers Analytics Engine ↗ with a major expansion of our SQL function library.

Workers Analytics Engine allows you to ingest and store high-cardinality data at scale (such as custom analytics) and query your data through a simple SQL API.

Today, we've expanded Workers Analytics Engine's SQL capabilities with several new functions:

New aggregate functions: ↗

• argMin() - Returns the value associated with the minimum in a group
• argMax() - Returns the value associated with the maximum in a group
• topK() - Returns an array of the most frequent values in a group
• topKWeighted() - Returns an array of the most frequent values in a group using weights
• first_value() - Returns the first value in an ordered set of values within a partition
• last_value() - Returns the last value in an ordered set of values within a partition

New bit functions: ↗

• bitAnd() - Returns the bitwise AND of two expressions
• bitCount() - Returns the number of bits set to one in the binary representation of a number
• bitHammingDistance() - Returns the number of bits that differ between two numbers
• bitNot() - Returns a number with all bits flipped
• bitOr() - Returns the inclusive bitwise OR of two expressions
• bitRotateLeft() - Rotates all bits in a number left by specified positions
• bitRotateRight() - Rotates all bits in a number right by specified positions
• bitShiftLeft() - Shifts all bits in a number left by specified positions
• bitShiftRight() - Shifts all bits in a number right by specified positions
• bitTest() - Returns the value of a specific bit in a number
• bitXor() - Returns the bitwise exclusive-or of two expressions

New mathematical functions: ↗

• abs() - Returns the absolute value of a number
• log() - Computes the natural logarithm of a number
• round() - Rounds a number to a specified number of decimal places
• ceil() - Rounds a number up to the nearest integer
• floor() - Rounds a number down to the nearest integer
• pow() - Returns a number raised to the power of another number

New string functions: ↗

• lowerUTF8() - Converts a string to lowercase using UTF-8 encoding
• upperUTF8() - Converts a string to uppercase using UTF-8 encoding

New encoding functions: ↗

• hex() - Converts a number to its hexadecimal representation
• bin() - Converts a string to its binary representation

New type conversion functions: ↗

• toUInt8() - Converts any numeric expression, or expression resulting in a string representation of a decimal, into an unsigned 8 bit integer

Whether you're building usage-based billing systems, customer analytics dashboards, or other custom analytics, these functions let you get the most out of your data. Get started with Workers Analytics Engine and explore all available functions in our SQL reference documentation.

New instance types provide up to 4 vCPU, 12 GiB of memory, and 20 GB of disk per container instance.

The dev and standard instance types are preserved for backward compatibility and are aliases for lite and standard-1, respectively. The standard-1 instance type now provides up to 8 GB of disk instead of only 4 GB.

See the getting started guide to deploy your first Container, and the limits documentation for more details on the available instance types and limits.

The ctx.exports API contains automatically-configured bindings corresponding to your Worker's top-level exports. For each top-level export extending WorkerEntrypoint, ctx.exports will contain a Service Binding by the same name, and for each export extending DurableObject (and for which storage has been configured via a migration), ctx.exports will contain a Durable Object namespace binding. This means you no longer have to configure these bindings explicitly in wrangler.jsonc/wrangler.toml.

Example:

import { WorkerEntrypoint } from "cloudflare:workers";

export class Greeter extends WorkerEntrypoint {
  greet(name) {
    return `Hello, ${name}!`;
  }
}

export default {
  async fetch(request, env, ctx) {
    let greeting = await ctx.exports.Greeter.greet("World")
    return new Response(greeting);
  }
}

At present, you must use the enable_ctx_exports compatibility flag to enable this API, though it will be on by default in the future.

See the API reference for more information.

Today, we're launching the new Cloudflare Pipelines: a streaming data platform that ingests events, transforms them with SQL, and writes to R2 as Apache Iceberg ↗ tables or Parquet files.

Pipelines can receive events via HTTP endpoints or Worker bindings, transform them with SQL, and deliver to R2 with exactly-once guarantees. This makes it easy to build analytics-ready warehouses for server logs, mobile application events, IoT telemetry, or clickstream data without managing streaming infrastructure.

For example, here's a pipeline that ingests clickstream events and filters out bot traffic while extracting domain information:

INSERT into events_table
SELECT
  user_id,
  lower(event) AS event_type,
  to_timestamp_micros(ts_us) AS event_time,
  regexp_match(url, '^https?://([^/]+)')[1]  AS domain,
  url,
  referrer,
  user_agent
FROM events_json
WHERE event = 'page_view'
  AND NOT regexp_like(user_agent, '(?i)bot|spider');

Get started by creating a pipeline in the dashboard or running a single command in Wrangler:

npx wrangler pipelines setup

Check out our getting started guide to learn how to create a pipeline that delivers events to an Iceberg table you can query with R2 SQL. Read more about today's announcement in our blog post ↗.

Today, we're launching the open beta for R2 SQL: A serverless, distributed query engine that can efficiently analyze petabytes of data in Apache Iceberg ↗ tables managed by R2 Data Catalog.

R2 SQL is ideal for exploring analytical and time-series data stored in R2, such as logs, events from Pipelines, or clickstream and user behavior data.

If you already have a table in R2 Data Catalog, running queries is as simple as:

npx wrangler r2 sql query YOUR_WAREHOUSE "
SELECT
    user_id,
    event_type,
    value
FROM events.user_events
WHERE event_type = 'CHANGELOG' or event_type = 'BLOG'
  AND __ingest_ts > '2025-09-24T00:00:00Z'
ORDER BY __ingest_ts DESC
LIMIT 100"

To get started with R2 SQL, check out our getting started guide or learn more about supported features in the SQL reference. For a technical deep dive into how we built R2 SQL, read our blog post ↗.

We’re shipping three updates to Browser Rendering:

• Playwright support is now Generally Available and synced with Playwright v1.55 ↗, giving you a stable foundation for critical automation and AI-agent workflows.
• We’re also adding Stagehand support (Beta) so you can combine code with natural language instructions to build more resilient automations.
• Finally, we’ve tripled limits for paid plans across both the REST API and Browser Sessions to help you scale.

To get started with Stagehand, refer to the Stagehand example that uses Stagehand and Workers AI to search for a movie on this example movie directory ↗, extract its details using natural language (title, year, rating, duration, and genre), and return the information along with a screenshot of the webpage.

const stagehand = new Stagehand({
	env: "LOCAL",
	localBrowserLaunchOptions: { cdpUrl: endpointURLString(env.BROWSER) },
	llmClient: new WorkersAIClient(env.AI),
	verbose: 1,
});

await stagehand.init();
const page = stagehand.page;

await page.goto("https://demo.playwright.dev/movies");

// if search is a multi-step action, stagehand will return an array of actions it needs to act on
const actions = await page.observe('Search for "Furiosa"');
for (const action of actions) await page.act(action);

await page.act("Click the search result");

// normal playwright functions work as expected
await page.waitForSelector(".info-wrapper .cast");

let movieInfo = await page.extract({
	instruction: "Extract movie information",
	schema: z.object({
		title: z.string(),
		year: z.number(),
		rating: z.number(),
		genres: z.array(z.string()),
		duration: z.number().describe("Duration in minutes"),
	}),
});

await stagehand.close();

AutoRAG is now AI Search! The new name marks a new and bigger mission: to make world-class search infrastructure available to every developer and business.

With AI Search you can now use models from different providers like OpenAI and Anthropic. By attaching your provider keys to the AI Gateway linked to your AI Search instance, you can use many more models for both embedding and inference.

To use AI Search with other model providers:

1. Add provider keys to AI Gateway
   1. Go to AI > AI Gateway in the dashboard.
   2. Select or create an AI gateway.
   3. In Provider Keys, choose your provider, click Add, and enter the key.
2. Connect a gateway to AI Search: When creating a new AI Search, select the AI Gateway with your provider keys. For an existing AI Search, go to Settings and switch to a gateway that has your keys under Resources.
3. Select models: Embedding models are only available to be changed when creating a new AI Search. Generation model can be selected when creating a new AI Search and can be changed at any time in Settings.

Once configured, your AI Search instance will be able to reference models available through your AI Gateway when making a /ai-search request:

export default {
  async fetch(request, env) {
    
    // Query your AI Search instance with a natural language question to an OpenAI model
    const result = await env.AI.autorag("my-ai-search").aiSearch({
      query: "What's new for Cloudflare Birthday Week?",
      model: "openai/gpt-5"
    });

    // Return only the generated answer as plain text
    return new Response(result.response, {
      headers: { "Content-Type": "text/plain" },
    });
  },
};

In the coming weeks we will also roll out updates to align the APIs with the new name. The existing APIs will continue to be supported for the time being. Stay tuned to the AI Search Changelog and Discord ↗ for more updates!

You can now run more Containers concurrently with higher limits on CPU, memory, and disk.

You can now run 1000 instances of the dev instance type, 400 instances of basic, or 100 instances of standard concurrently.

This opens up new possibilities for running larger-scale workloads on Containers.

See the getting started guide to deploy your first Container, and the limits documentation for more details on the available instance types and limits.

You can now enable automatic compaction for Apache Iceberg ↗ tables in R2 Data Catalog to improve query performance.

Compaction is the process of taking a group of small files and combining them into fewer larger files. This is an important maintenance operation as it helps ensure that query performance remains consistent by reducing the number of files that needs to be scanned.

To enable automatic compaction in R2 Data Catalog, find it under R2 Data Catalog in your R2 bucket settings in the dashboard.

Or with Wrangler, run:

npx wrangler r2 bucket catalog compaction enable <BUCKET_NAME>  --target-size 128 --token <API_TOKEN>

To get started with compaction, check out manage catalogs. For best practices and limitations, refer to about compaction.

You can run multiple Workers in a single dev command by passing multiple config files to wrangler dev:

wrangler dev --config ./web/wrangler.jsonc --config ./api/wrangler.jsonc

Previously, if you ran the command above and then also ran wrangler dev for a different Worker, the Workers running in separate wrangler dev sessions could not communicate with each other. This prevented you from being able to use Service Bindings ↗ and Tail Workers ↗ in local development, when running separate wrangler dev sessions.

Now, the following works as expected:

# Terminal 1: Run your application that includes both Web and API workers
wrangler dev --config ./web/wrangler.jsonc --config ./api/wrangler.jsonc

# Terminal 2: Run your auth worker separately
wrangler dev --config ./auth/wrangler.jsonc

These Workers can now communicate with each other across separate dev commands, regardless of your development setup.

export default {
	async fetch(request, env) {
		// This service binding call now works across dev commands
		const authorized = await env.AUTH.isAuthorized(request);

		if (!authorized) {
			return new Response("Unauthorized", { status: 401 });
		}

		return new Response("Hello from API Worker!", { status: 200 });
	},
};

Check out the Developing with multiple Workers guide to learn more about the different approaches and when to use each one.

AutoRAG now includes a Metrics tab that shows how your data is indexed and searched. Get a clear view of the health of your indexing pipeline, compare usage between ai-search and search, and see which files are retrieved most often.

You can find these metrics within each AutoRAG instance:

• Indexing: Track how files are ingested and see status changes over time.
• Search breakdown: Compare usage between ai-search and search endpoints.
• Top file retrievals: Identify which files are most frequently retrieved in a given period.

Try it today in AutoRAG.

Rate Limiting within Cloudflare Workers is now Generally Available (GA).

The ratelimit binding is now stable and recommended for all production workloads. Existing deployments using the unsafe binding will continue to function to allow for a smooth transition.

For more details, refer to Workers Rate Limiting documentation.

In workers-rs ↗, Rust panics were previously non-recoverable. A panic would put the Worker into an invalid state, and further function calls could result in memory overflows or exceptions.

Now, when a panic occurs, in-flight requests will throw 500 errors, but the Worker will automatically and instantly recover for future requests.

This ensures more reliable deployments. Automatic panic recovery is enabled for all new workers-rs deployments as of version 0.6.5, with no configuration required.

Rust Workers are built with Wasm Bindgen, which treats panics as non-recoverable. After a panic, the entire Wasm application is considered to be in an invalid state.

We now attach a default panic handler in Rust:

std::panic::set_hook(Box::new(move |panic_info| {
  hook_impl(panic_info);
}));

Which is registered by default in the JS initialization:

import { setPanicHook } from "./index.js";
setPanicHook(function (err) {
	console.error("Panic handler!", err);
});

When a panic occurs, we reset the Wasm state to revert the Wasm application to how it was when the application started.

We worked upstream on the Wasm Bindgen project to implement a new --experimental-reset-state-function compilation option ↗ which outputs a new __wbg_reset_state function.

This function clears all internal state related to the Wasm VM, and updates all function bindings in place to reference the new WebAssembly instance.

One other necessary change here was associating Wasm-created JS objects with an instance identity. If a JS object created by an earlier instance is then passed into a new instance later on, a new "stale object" error is specially thrown when using this feature.

Building on this new Wasm Bindgen feature, layered with our new default panic handler, we also added a proxy wrapper to ensure all top-level exported class instantiations (such as for Rust Durable Objects) are tracked and fully reinitialized when resetting the Wasm instance. This was necessary because the workerd runtime will instantiate exported classes, which would then be associated with the Wasm instance.

This approach now provides full panic recovery for Rust Workers on subsequent requests.

Of course, we never want panics, but when they do happen they are isolated and can be investigated further from the error logs - avoiding broader service disruption.

In the future, full support for recoverable panics could be implemented without needing reinitialization at all, utilizing the WebAssembly Exception Handling ↗ proposal, part of the newly announced WebAssembly 3.0 ↗ specification. This would allow unwinding panics as normal JS errors, and concurrent requests would no longer fail.

We're making significant improvements to the reliability of Rust Workers ↗. Join us in #rust-on-workers on the Cloudflare Developers Discord ↗ to stay updated.

You can now route private traffic to Cloudflare Tunnel based on a hostname or domain, moving beyond the limitations of IP-based routing. This new capability is free for all Cloudflare One customers.

Previously, Tunnel routes could only be defined by IP address or CIDR range. This created a challenge for modern applications with dynamic or ephemeral IP addresses, often forcing administrators to maintain complex and brittle IP lists.

What’s new:

• Hostname & Domain Routing: Create routes for individual hostnames (e.g., payroll.acme.local) or entire domains (e.g., *.acme.local) and direct their traffic to a specific Tunnel.
• Simplified Zero Trust Policies: Build resilient policies in Cloudflare Access and Gateway using stable hostnames, making it dramatically easier to apply per-resource authorization for your private applications.
• Precise Egress Control: Route traffic for public hostnames (e.g., bank.example.com) through a specific Tunnel to enforce a dedicated source IP, solving the IP allowlist problem for third-party services.
• No More IP Lists: This feature makes the workaround of maintaining dynamic IP Lists for Tunnel connections obsolete.

Get started in the Tunnels section of the Zero Trust dashboard with your first private hostname or public hostname route.

Learn more in our blog post ↗.

We recently increased the available disk space from 8 GB to 20 GB for all plans. Building on that improvement, we’re now doubling the CPU power available for paid plans — from 2 vCPU to 4 vCPU.

These changes continue our focus on making Workers Builds faster and more reliable.

• Fast build times: Even single-threaded workloads benefit from having more vCPUs
• 2x faster multi-threaded builds: Tools like esbuild ↗ and webpack ↗ can now utilize additional cores, delivering near-linear performance scaling

All other build limits — including memory, build minutes, and timeout remain unchanged.

To prevent the accidental exposure of applications, we've updated how Worker preview URLs (<PREVIEW>-<WORKER_NAME>.<SUBDOMAIN>.workers.dev) are handled. We made this change to ensure preview URLs are only active when intentionally configured, improving the default security posture of your Workers.

We performed a one-time update to disable preview URLs for existing Workers where the workers.dev subdomain was also disabled.

Because preview URLs were historically enabled by default, users who had intentionally disabled their workers.dev route may not have realized their Worker was still accessible at a separate preview URL. This update was performed to ensure that using a preview URL is always an intentional, opt-in choice.

If your Worker was affected, its preview URL (<PREVIEW>-<WORKER_NAME>.<SUBDOMAIN>.workers.dev) will now direct to an informational page explaining this change.

How to Re-enable Your Preview URL

If your preview URL was disabled, you can re-enable it via the Cloudflare dashboard by navigating to your Worker's Settings page and toggling on the Preview URL.

Alternatively, you can use Wrangler by adding the preview_urls = true setting to your Wrangler file and redeploying the Worker.

{
  "preview_urls": true
}

preview_urls = true

Note: You can set preview_urls = true with any Wrangler version that supports the preview URL flag (v3.91.0+). However, we recommend updating to v4.34.0 or newer, as this version defaults preview_urls to false, ensuring preview URLs are always enabled by explicit choice.

Three months ago we announced the public beta of remote bindings for local development. Now, we're excited to say that it's available for everyone in Wrangler, Vite, and Vitest without using an experimental flag!

With remote bindings, you can now connect to deployed resources like R2 buckets and D1 databases while running Worker code on your local machine. This means you can test your local code changes against real data and services, without the overhead of deploying for each iteration.

To enable remote bindings, add "remote" : true to each binding that you want to rely on a remote resource running on Cloudflare:

{
	"name": "my-worker",
	// Set this to today's date
	"compatibility_date": "2026-08-16",

	"r2_buckets": [
		{
			"bucket_name": "screenshots-bucket",
			"binding": "screenshots_bucket",
			"remote": true,
		},
	],
}

name = "my-worker"
# Set this to today's date
compatibility_date = "2026-08-16"

[[r2_buckets]]
bucket_name = "screenshots-bucket"
binding = "screenshots_bucket"
remote = true

When remote bindings are configured, your Worker still executes locally, but all binding calls are proxied to the deployed resource that runs on Cloudflare's network.

You can try out remote bindings for local development today with:

• Wrangler v4.37.0
• The Cloudflare Vite Plugin
• The Cloudflare Vitest Plugin

D1 now detects read-only queries and automatically attempts up to two retries to execute those queries in the event of failures with retryable errors. You can access the number of execution attempts in the returned response metadata property total_attempts.

At the moment, only read-only queries are retried, that is, queries containing only the following SQLite keywords: SELECT, EXPLAIN, WITH. Queries containing any SQLite keyword ↗ that leads to database writes are not retried.

The retry success ratio among read-only retryable errors varies from 5% all the way up to 95%, depending on the underlying error and its duration (like network errors or other internal errors).

The retry success ratio among all retryable errors is lower, indicating that there are write-queries that could be retried. Therefore, we recommend D1 users to continue applying retries in their own code for queries that are not read-only but are idempotent according to the business logic of the application.

D1 ensures that any retry attempt does not cause database writes, making the automatic retries safe from side-effects, even if a query causing changes slips through the read-only detection. D1 achieves this by checking for modifications after every query execution, and if any write occurred due to a retry attempt, the query is rolled back.

The read-only query detection heuristics are simple for now, and there is room for improvement to capture more cases of queries that can be retried, so this is just the beginning.

The number of recent versions available for a Worker rollback has been increased from 10 to 100.

This allows you to:

• Promote any of the 100 most recent versions to be the active deployment.

• Split traffic using gradual deployments between your latest code and any of the 100 most recent versions.

You can do this through the Cloudflare dashboard or with Wrangler's rollback command

Learn more about versioned deployments and rollbacks.

We've shipped a new release for the Agents SDK ↗ bringing full compatibility with AI SDK v5 ↗ and introducing automatic message migration that handles all legacy formats transparently.

This release includes improved streaming and tool support, tool confirmation detection (for "human in the loop" systems), enhanced React hooks with automatic tool resolution, improved error handling for streaming responses, and seamless migration utilities that work behind the scenes.

This makes it ideal for building production AI chat interfaces with Cloudflare Workers AI models, agent workflows, human-in-the-loop systems, or any application requiring reliable message handling across SDK versions — all while maintaining backward compatibility.

Additionally, we've updated workers-ai-provider v2.0.0, the official provider for Cloudflare Workers AI models, to be compatible with AI SDK v5.

Creates a new chat interface with enhanced v5 capabilities.

// Basic chat setup
const { messages, sendMessage, addToolResult } = useAgentChat({
	agent,
	experimental_automaticToolResolution: true,
	tools,
});

// With custom tool confirmation
const chat = useAgentChat({
	agent,
	experimental_automaticToolResolution: true,
	toolsRequiringConfirmation: ["dangerousOperation"],
});

Tools are automatically categorized based on their configuration:

const tools = {
	// Auto-executes (has execute function)
	getLocalTime: {
		description: "Get current local time",
		inputSchema: z.object({}),
		execute: async () => new Date().toLocaleString(),
	},

	// Requires confirmation (no execute function)
	deleteFile: {
		description: "Delete a file from the system",
		inputSchema: z.object({
			filename: z.string(),
		}),
	},

	// Server-executed (no client confirmation)
	analyzeData: {
		description: "Analyze dataset on server",
		inputSchema: z.object({ data: z.array(z.number()) }),
		serverExecuted: true,
	},
} satisfies Record<string, AITool>;

Send messages using the new v5 format with parts array:

// Text message
sendMessage({
	role: "user",
	parts: [{ type: "text", text: "Hello, assistant!" }],
});

// Multi-part message with file
sendMessage({
	role: "user",
	parts: [
		{ type: "text", text: "Analyze this image:" },
		{ type: "image", image: imageData },
	],
});

Simplified logic for detecting pending tool confirmations:

const pendingToolCallConfirmation = messages.some((m) =>
	m.parts?.some(
		(part) => isToolUIPart(part) && part.state === "input-available",
	),
);

// Handle tool confirmation
if (pendingToolCallConfirmation) {
	await addToolResult({
		toolCallId: part.toolCallId,
		tool: getToolName(part),
		output: "User approved the action",
	});
}

Seamlessly handle legacy message formats without code changes.

// All these formats are automatically converted:

// Legacy v4 string content
const legacyMessage = {
	role: "user",
	content: "Hello world",
};

// Legacy v4 with tool calls
const legacyWithTools = {
	role: "assistant",
	content: "",
	toolInvocations: [
		{
			toolCallId: "123",
			toolName: "weather",
			args: { city: "SF" },
			state: "result",
			result: "Sunny, 72°F",
		},
	],
};

// Automatically becomes v5 format:
// {
//   role: "assistant",
//   parts: [{
//     type: "tool-call",
//     toolCallId: "123",
//     toolName: "weather",
//     args: { city: "SF" },
//     state: "result",
//     result: "Sunny, 72°F"
//   }]
// }

Migrate tool definitions to use the new inputSchema property.

// Before (AI SDK v4)
const tools = {
	weather: {
		description: "Get weather information",
		parameters: z.object({
			city: z.string(),
		}),
		execute: async (args) => {
			return await getWeather(args.city);
		},
	},
};

// After (AI SDK v5)
const tools = {
	weather: {
		description: "Get weather information",
		inputSchema: z.object({
			city: z.string(),
		}),
		execute: async (args) => {
			return await getWeather(args.city);
		},
	},
};

Seamless integration with Cloudflare Workers AI models through the updated workers-ai-provider v2.0.0.

Use Cloudflare Workers AI models directly in your agent workflows:

import { createWorkersAI } from "workers-ai-provider";
import { useAgentChat } from "agents/ai-react";

// Create Workers AI model (v2.0.0 - same API, enhanced v5 internals)
const model = createWorkersAI({
	binding: env.AI,
})("@cf/meta/llama-3.2-3b-instruct");

Workers AI models now support v5 file handling with automatic conversion:

// Send images and files to Workers AI models
sendMessage({
	role: "user",
	parts: [
		{ type: "text", text: "Analyze this image:" },
		{
			type: "file",
			data: imageBuffer,
			mediaType: "image/jpeg",
		},
	],
});

// Workers AI provider automatically converts to proper format

Enhanced streaming support with automatic warning detection:

// Streaming with Workers AI models
const result = await streamText({
	model: createWorkersAI({ binding: env.AI })("@cf/meta/llama-3.2-3b-instruct"),
	messages,
	onChunk: (chunk) => {
		// Enhanced streaming with warning handling
		console.log(chunk);
	},
});

Update your imports to use the new v5 types:

// Before (AI SDK v4)
import type { Message } from "ai";
import { useChat } from "ai/react";

// After (AI SDK v5)
import type { UIMessage } from "ai";
// or alias for compatibility
import type { UIMessage as Message } from "ai";
import { useChat } from "@ai-sdk/react";

• Migration Guide ↗ - Comprehensive migration documentation
• AI SDK v5 Documentation ↗ - Official AI SDK migration guide
• An Example PR showing the migration from AI SDK v4 to v5 ↗
• GitHub Issues ↗ - Report bugs or request features

We'd love your feedback! We're particularly interested in feedback on:

• Migration experience - How smooth was the upgrade process?
• Tool confirmation workflow - Does the new automatic detection work as expected?
• Message format handling - Any edge cases with legacy message conversion?