checked with the team: They can use `keep_alive` , eg `const browser = await puppeteer.launch(env.M

checked with the team:

They can use

keep_alive

keep_alive

keep_alive

keep_alive , eg

const browser = await puppeteer.launch(env.MYBROWSER, { keep_alive: 600000 });

const browser = await puppeteer.launch(env.MYBROWSER, { keep_alive: 600000 });

const browser = await puppeteer.launch(env.MYBROWSER, { keep_alive: 600000 });

const browser = await puppeteer.launch(env.MYBROWSER, { keep_alive: 600000 }); to keep the browser alive for 10 minutes, and then, in order to get a valid

sessionId

sessionId

sessionId

sessionId, they need to call

https://developers.cloudflare.com/browser-rendering/platform/puppeteer/#list-open-sessions

https://developers.cloudflare.com/browser-rendering/platform/puppeteer/#list-open-sessions

https://developers.cloudflare.com/browser-rendering/platform/puppeteer/#list-open-sessions

https://developers.cloudflare.com/browser-rendering/platform/puppeteer/#list-open-sessions .

Full example here https://developers.cloudflare.com/browser-rendering/workers-bindings/reuse-sessions/#4-code

Cloudflare Docs

Reuse sessions

The best way to improve the performance of your browser rendering Worker is to reuse sessions. One way to do that is via Durable Objects, which allows you to keep a long running connection from a Worker to a browser. Another way is to keep the browser open after you've finished with it, and connect to that session each time you have a new request.

KKathy checked with the team: They can use `keep_alive` , eg `const browser = await pu...

Celestial Rose•11/4/25, 11:48 PM

Hey! Thanks for your reply, I have found this link already, but does this work for stagehand ? This seems to only work for puppeteer; I literally posted this link in my OG post haha

Vikash•11/5/25, 4:29 AM

The /snapshot API documentation is outdated? or the response has been changed, I don't see

status

status

in response, instead the response just display success: true/falsesuccess: true/false, while I was expecting the response status: 2xx, 4xx etc retuned by given URL. As it always says success: true, even for 404 link as well

Vikash•11/5/25, 4:32 AM

For example, I passed a bad link in url - which return 404 on browser, but there are no way to find out it retuned 404 using the /snapshot API. Which enforce me to send 2 requests (one for status, another for snapshot)

Nathan•11/5/25, 7:28 PM

Have anyone ran into an issue while trying to download a file via Playwright Download.createReadStream()Download.createReadStream() API?

The readable has no data in it. Doesn't throw an error, but just 0 chunks.

KKathy you can `addStyleTag` to customize. see example here https://developers.cloudfla...

aem•11/7/25, 2:37 AM

Yes we've tried both addStyleTagaddStyleTag as well as an embedded stylesheet in the HTML with fontfont as an allowed resource type, neither work

Mazin•11/9/25, 8:12 PM

i made a worker for browser rendering and deployed it, the first request i made i got this error, why?

Error: Unable to create new browser: code: 429: message: Rate limit exceeded

Stack trace:
Error: Unable to create new browser: code: 429: message: Rate limit exceeded
at PuppeteerWorkers.acquire (index.js:18328:13)
at async PuppeteerWorkers.launch (index.js:18238:22)
at async Object.fetch (index.js:19976:23)

Mazin•11/9/25, 8:36 PM

anyone?

Masquerades•11/10/25, 12:48 AM

Hey team, what's going on here?

I'm trying to run the documented code snippet from https://developers.cloudflare.com/browser-rendering/rest-api/json-endpoint/#using-a-custom-model-byo-api-key

Simply replacing the ai provider from anthropic to google-ai-studio/gemini-2.5-flash, or google-ai-studio/gemini-flash-latest. I get the error {"success":false,"errors":[{"message":"AI error: undefined: undefined"}]}%. Looks like someone else is having the problem too: https://community.cloudflare.com/t/browser-rendering-json-with-custom-ai/854912

Am I missing something here, or is the snippet in the documentation incorrect?

zapalona_swieczka•11/11/25, 7:58 AM

any chance anyone is making screenshot of an element? screenshot of playwright page works excelent, but when I try to run locator to find some element and then run screenshot - timeouts, tried like 5-6 different methods to get element

zapalona_swieczka•11/12/25, 10:54 AM

I wonder how much of time of browser rendering cpu time of workers is counted. Since browser rendering is paid, I hope majority of time spent on that will not be billed as worker cpu-time?

MMazin i made a worker for browser rendering and deployed it, the first request i made ...

mtnrunner•11/12/25, 7:31 PM

did you find a resolution for this? same thing with me

mtnrunner•11/12/25, 8:31 PM

Alas, i assume there is a bug

  {
    "sessionId": "31ab6f51-4628-41e0-bf91-477cb48fda4a",
    "startTime": 1762969852952,
    "endTime": 1762969856860,
    "closeReason": 12,
    "closeReasonText": "UsedBrowserTimeLimitExceeded"
  }

  {
    "sessionId": "31ab6f51-4628-41e0-bf91-477cb48fda4a",
    "startTime": 1762969852952,
    "endTime": 1762969856860,
    "closeReason": 12,
    "closeReasonText": "UsedBrowserTimeLimitExceeded"
  }

that's about 4 seconds and i see in the UI it looks like

31ab6f51-4628-41e0-bf91-477cb48fda4a
Dec 31, 1969 5:00:00 PM
Nov 12, 2025 10:50:56 AM

31ab6f51-4628-41e0-bf91-477cb48fda4a
Dec 31, 1969 5:00:00 PM
Nov 12, 2025 10:50:56 AM

Mmtnrunner Alas, i assume there is a bug ``` { "sessionId": "31ab6f51-4628-41e0-bf91-...

KathyOP•11/12/25, 9:04 PM

Are you on the Workers Paid plan? or Workers Free plan?

Users on Workers Free are limited to 10 minutes a day https://developers.cloudflare.com/browser-rendering/limits/#workers-free

Cloudflare Docs

Limits

Learn about the limits associated with Browser Rendering.

Mmtnrunner Alas, i assume there is a bug ``` { "sessionId": "31ab6f51-4628-41e0-bf91-...

KathyOP•11/12/25, 9:05 PM

timestart being the wrong year is a bug - on our list to fix

KKathy timestart being the wrong year is a bug - on our list to fix

mtnrunner•11/12/25, 9:55 PM

awesome! its the free plan, am planning to upgrade but just wanted to test this out - this is the only request i made today and everything after said it was 429'd - so 4 seconds == hit my limit for the day based on the logs i see

Mmtnrunner awesome! its the free plan, am planning to upgrade but just wanted to test this ...

KathyOP•11/13/25, 12:41 AM

can you dm me your account id?

odd you weren't able to use 10 minutes

KathyOP•11/13/25, 8:39 PM

Hi y'all!

Browser Rendering is due for a dashboard update

What would you like to see added to the dashboard?

Please feel free to chime in here on chat or we can discuss live! https://calendar.app.google/Xe4EWuvaAVvehDPC6

Browser Rendering Chat - Kathy Liao

Mmtnrunner awesome! its the free plan, am planning to upgrade but just wanted to test this ...

KathyOP•11/14/25, 5:06 PM

this is fixed! thanks for raising~

MMasquerades Hey team, what's going on here? I'm trying to run the documented code snippet f...

TheCrazyBob•11/16/25, 1:05 PM

@Kathy I am having the same issue on my end. I can see another user reporting this as well: https://community.cloudflare.com/t/browser-rendering-json-with-custom-ai/854912

1984 Ford Laser•11/18/25, 6:16 AM

Have tried using the Browser Rendering REST API, but it isn't working. The Workers API binding works fine, and I have created a new API token with Browser Rendering Edit perms. I can confirm that using a different token gives an auth error. Doesn't work with either a URL or HTML body param

1984 Ford Laser•11/18/25, 6:17 AM

Also, in the Workers binding, is it normal for it to take 7-10 seconds to get a screenshot response for a very simple HTML page hosted on CF?

MMasquerades Hey team, what's going on here? I'm trying to run the documented code snippet f...

TheCrazyBob•11/18/25, 10:32 AM

Were you able to find a solution? The custom AI provider feature doesn't work

arnaseef•11/20/25, 6:08 PM

https://community.cloudflare.com/t/browser-rendering-json-with-custom-ai/854912/2

@Kathy

KathyOP•11/21/25, 6:21 PM

json endpoint bugs have been fixed for both gemini and openai! thanks for reporting @arnaseef , @TheCrazyBob , @Masquerades

KKathy json endpoint bugs have been fixed for both gemini and openai! thanks for report...

TheCrazyBob•11/22/25, 3:37 PM

I can't get it to work with openai's models. Haven't tried gemini yet. I always get the following error:

{
  "success": false,
  "errors": [
    {
      "message": "Unable to form JSON based on user prompt and/or webpage text. Please try with a different prompt try passing a response_format."
    }
  ]
}

{
  "success": false,
  "errors": [
    {
      "message": "Unable to form JSON based on user prompt and/or webpage text. Please try with a different prompt try passing a response_format."
    }
  ]
}

hunterleung.•11/27/25, 5:02 PM

Hi, could anybody give a sample usage about the snapshot API (https://developers.cloudflare.com/api/python/resources/browser_rendering/subresources/snapshot/methods/create/) ?
I wanna add the "full_page" and "type" to request, but I always get a error:

I have tried these 2 request body ,but not work either:

{
    "url": "https://www.bbc.com/",
    "screenshot_options":  {"full_page" : true , "type":"webp" }
  }

{
    "url": "https://www.bbc.com/",
    "screenshot_options":  {"full_page" : true , "type":"webp" }
  }

{
    "url": "https://www.bbc.com/",
    "full_page" : true , 
    "type":"webp" }
  }

{
    "url": "https://www.bbc.com/",
    "full_page" : true , 
    "type":"webp" }
  }

I got this error:

{
    "errors": [
        {
            "code": "unrecognized_keys",
            "keys": [
                "screenshot_options"
            ],
            "path": [
                "body"
            ],
            "message": "Unrecognized key(s) in object: 'screenshot_options'"
        }
    ],
    "success": false,
    "result": {}
}

{
    "errors": [
        {
            "code": "unrecognized_keys",
            "keys": [
                "screenshot_options"
            ],
            "path": [
                "body"
            ],
            "message": "Unrecognized key(s) in object: 'screenshot_options'"
        }
    ],
    "success": false,
    "result": {}
}

How should I fix it ? Thanks

Cloudflare API | Browser Rendering › Snapshot › create

Interact with Cloudflare's products and services via the Cloudflare API

hunterleung.•11/28/25, 3:31 AM

@Kathy hi,Kathy , could you help me ?

Hhunterleung.Hi, could anybody give a sample usage about the snapshot API (https://developer...

Gibbeth•11/28/25, 3:35 AM

you're running POST @ /accounts/{account_id}/browser-rendering/snapshot right?

Gibbeth•11/28/25, 3:36 AM

based on the docs ur i think the first example should be the functioning one

GGibbeth you're running POST @ /accounts/{account_id}/browser-rendering/snapshot right?

hunterleung.•11/28/25, 4:10 AM

I see, what happend. "screenshot_options" should be "screenshotOptions"

Hhunterleung.I see, what happend. "screenshot_options" should be "screenshotOptions"

Gibbeth•11/28/25, 4:57 AM

huh, that's weird. that wasn't very clear in the docs

Gibbeth•11/28/25, 4:58 AM

i was super confused bc to me scerenshot_options should've worked

Hhunterleung.I see, what happend. "screenshot_options" should be "screenshotOptions"

KathyOP•12/1/25, 6:03 PM

think issue might have been you were in python docs so you got

screenshot_options

screenshot_options

instead of

screenshotOptions

screenshotOptions

Ccosmo Hello! I'm testing this amazing API and I just found a bug. My total usage is l...

KathyOP•12/1/25, 6:03 PM

can you dm me your account id

welz•12/4/25, 3:52 PM

What does this mean?

Unable to connect to existing session <uuid> (it may still be in use or not ready yet) - retry or launch a new browser: TargetCloseError: Protocol error (Browser.getVersion): Target closed

Unable to connect to existing session <uuid> (it may still be in use or not ready yet) - retry or launch a new browser: TargetCloseError: Protocol error (Browser.getVersion): Target closed

I'm seeing this periodically (and inconsistently) on browser launch.

Wwelz What does this mean? ``` Unable to connect to existing session <uuid> (it may st...

KathyOP•12/4/25, 9:09 PM

from a quick perusal, we don't see that error in our logs. are you sure that's from us?

KKathy from a quick perusal, we don't see that error in our logs. are you sure that's f...

welz•12/5/25, 8:42 AM

It's definitely coming from Browser Rendering - the uuid also matches up with my BR logs. I'm invoking using

await puppeteer.launch(env.BROWSER)

await puppeteer.launch(env.BROWSER)

so it shouldn't even be trying to connect to an existing session.

TheCrazyBob•12/6/25, 8:30 PM

The

cacheTTL

cacheTTL

parameter doesn't seem to work. The request still gets cached, and I keep getting the same response even after waiting for a few minutes. Can someone look into this?

1984 Ford Laser•12/9/25, 1:14 AM

Im still having issues accessing the Browser Rendering REST API, every request returns a

Not Found.

Not Found.

response. I have Workers Paid and have tried both an Account API Token and User API Token with Edit access to the Browser Rendering perm. Any ideas to try?

11984 Ford Laser Im still having issues accessing the Browser Rendering REST API, every request r...

KathyOP•12/9/25, 1:16 AM

dm me your account id please

KKathy dm me your account id please

1984 Ford Laser•12/9/25, 1:17 AM

Done

1984 Ford Laser•12/9/25, 1:23 AM

To add, I just set up a quick test Worker and the Worker Binding for this seems to work fine

Ray Li•12/15/25, 5:02 PM

Is browser rendering == CF container+ playwright?

checked with the team: They can use `keep_alive` , eg `const browser = await puppeteer.launch(env.M

Similar Threads

Similar Threads

Similar Threads