Is it known that AI Gateway with Unified Billing (Beta) doesn’t seem to work with OpenAI when using

Is it known that AI Gateway with Unified Billing (Beta) doesn’t seem to work with OpenAI when using the Vercel AI SDK together with Cloudflare’s ai-gateway-provider package?

In my testing:
• OpenAI → fails entirely through Unified Billing
• Google Generative AI → the non-streaming functions (generateText, generateObject) work, but the streaming variants do not
• Anthropic → both streaming and non-streaming work as expected

This seems odd, since the documentation states that Unified Billing supports all of these providers. Is this a known limitation or an issue on Cloudflare’s side?

I know it’s still beta but just want to know.

LLemmoTresto Is it known that AI Gateway with Unified Billing (Beta) doesn’t seem to work wit...

nwp•11/25/25, 4:30 PM

I hope you get an answer to this because I’m struggling with the same thing.

LemmoTrestoOP•11/25/25, 9:00 PM

Hoping on an answer from an employee, they might know more

Ccosbgn This works if anyone searches this: ```js const openai = createOpenAICompa...

nwp•11/26/25, 2:28 PM

Should be able to specify ‘openai/gpt-5’ or ‘groq/gpt-oss-120b’ as the model and it should work, but it doesn’t.

LemmoTrestoOP•11/26/25, 5:50 PM

I use the vercel ai sdk with a provider registry and do it like that instead. You can specify your provider/model that way.

LLemmoTresto Is it known that AI Gateway with Unified Billing (Beta) doesn’t seem to work wit...

LemmoTrestoOP•11/26/25, 5:50 PM

Still hoping for clarification on this

Victor•11/27/25, 5:15 AM

Is there an progress on supporting v1v1 and v1-previewv1-preview data plane API version (in azure) for ai gateway? Currently it breaks - unable to find resource (I assume url changes)

Victor•11/27/25, 5:16 AM

v1v1 is the new GA version for a while now and the azure adapter in cercel ai sdk now defaults to it as well, breaking any usage relying on defaults

nwp•11/27/25, 6:42 AM

I’m pretty close to giving up on Cloudflare and just paying $20 to Vercel.

Maxz•11/27/25, 1:32 PM

is cloudflare ai gateway down?
Why can I call some models but not others?
do I need to top up more ?

Maxz•11/27/25, 1:33 PM

{
"code": "Some resource has been exhausted",
"error": "Your team 23361c02-xxxx has either used all available credits or reached its monthly spending limit. To continue making API requests, please purchase more credits or raise your spending limit."
}

what is this team concept ? I don't have team.

watershed•11/30/25, 6:18 AM

*

Subject: Bug Report: Google Gemini File Upload Size Mismatch (via SDK Monkey Patch)

Description:
I am reporting a critical issue with Google Gemini (AI Studio) file uploads when routed through Cloudflare AI Gateway.

I am using the official google-genaigoogle-genai Python SDK. Since the SDK's resumable upload protocol automatically attempts to connect directly to generativelanguage.googleapis.comgenerativelanguage.googleapis.com for the data transfer phase (Step 2), I applied a monkey patch to the SDK to intercept the upload_urlupload_url and rewrite the hostname to my Cloudflare Gateway endpoint.

The Issue:
While the initial handshake succeeds, the actual file data transfer fails when routed through the Gateway. Google returns a 400 Bad Request400 Bad Request indicating that the data received is significantly larger (almost double) than the expected file size.

It seems the Gateway is mishandling the Transfer-Encoding: chunkedTransfer-Encoding: chunked stream sent by the Python httpxhttpx client, possibly causing data duplication or incorrect packet retries.

Error Log:
{ "code": 400, "message": "Can not finalize upload. Current size is 10085593. Expected final size is 5556955.", "status": "Bad Request" }
{ "code": 400, "message": "Can not finalize upload. Current size is 10085593. Expected final size is 5556955.", "status": "Bad Request" }

(Note: The Current sizeCurrent size ~10MB is nearly double the ExpectedExpected ~5.5MB.)

Reproduction Details:
SDK: Python google-genaigoogle-genai (using httpxhttpx).
Method: Monkey Patch applied to client._api_client.upload_fileclient._api_client.upload_file to replace googleapis.comgoogleapis.com with the Cloudflare Gateway URL.
Behavior: The SDK streams the file to Cloudflare -> Cloudflare forwards to Google -> Google rejects it due to size mismatch.

Reference Trace (CF-Ray):
9a6803f82f942a86-LAX9a6803f82f942a86-LAX

Could the team please investigate how the Gateway proxies chunked upload streams for the Google provider?

*

R•12/2/25, 6:53 AM

hello, im looking to set up logpush for cloudflare ai gateway to datadog following this guide: https://developers.cloudflare.com/ai-gateway/observability/logging/logpush/.

is there a way to send these logs without encryption? it seems like datadog does not offer a way to decrypt them, and that would mean that the logpush job that i've created for these logs to be sent to datadog would be useless since they can't be decrypted at the datadog level

Cloudflare Docs

Workers Logpush

AI Gateway allows you to securely export logs to an external storage location, where you can decrypt and process them.
You can toggle Workers Logpush on and off in the Cloudflare dashboard settings. This product is available on the Workers Paid plan. For pricing information, refer to Pricing.

LemmoTrestoOP•12/2/25, 6:09 PM

Has anyone used openrouter successfully with BYOK?

LemmoTrestoOP•12/2/25, 6:09 PM

Tells me it is missing authentication, using the vercel sdk, and the openrouter provider package.

LLaPh Although the output is set to be image-only, it's being treated as text generati...

Maxz•12/3/25, 8:58 AM

is the playground vide coded? so many bugs

Maxz•12/3/25, 8:59 AM

These unfinished pages are definitely damaging cf's brand.

Ssodepr I’ve seen several posts mentioning that Cloudflare AI Gateway routes requests to...

sarme•12/4/25, 9:39 AM

We're are getting the same issue. We have no server in Asia at all but all requests are getting redirected to one place which is restricted from open Ai without any control for placement from ourside.

{
"error": {
"code": "unsupported_country_region_territory",
"message": "Country, region, or territory not supported",
"param": null,
"type": "request_forbidden"
}
}

Has anyone found anyway to solve it?

MMaxz is cloudflare ai gateway down? Why can I call some models but not others? do I n...

sarme•12/4/25, 9:41 AM

Our Open AI requests are also redirected to one place which is restricted. I wonder if they are related.

Hharshil1712 forwarding this to the team.

sarme•12/4/25, 9:44 AM

This is happening again. Our error rate has jumped to 25%. Could you have a look again? @Kathy @harshil1712

Palash•12/8/25, 7:38 AM

@alex We have updated the AI SDK documentation page with examples of using the ai-gateway-providerai-gateway-provider along with examples of using AI SDK with BYOK/Unified Billing

https://developers.cloudflare.com/ai-gateway/integrations/vercel-ai-sdk/

Palash•12/9/25, 5:17 PM

Hey everyone

@Rishit Bansal and I work on the AI Gateway team. We’re catching up on some older threads now (sorry for the slow replies there!) and will be keeping a closer eye on this space. If you have any feedback or run into a bug, drop it here or tag us, we'll jump in as quickly as we can

LLaPh There are countless AI providers, and Google's image generation (Nano Banana ser...

Rishit Bansal•12/11/25, 5:00 AM

We are adding support for showing usage for more and more models/providers! There should be a fix releasing out sometime this week which will also add support for price calculation on embedding models on google ai studio/workers ai / vertex and other providers. If there's something specific which you immediately see that doesn't work: do let us know, and we'll add this in on a higher priority. But we also have some internal alerting setup which lets us know when usage calculation fails, which we keep an eye on!

Regarding the playground: yes, currently it only supports text generation models which specifically use the chat completions API format. We're working on limiting the list of models which shows on the playground to a smaller list, and also simultaneously seeing if we can have support for a few image generation/embeddings models there as well.

Jez•12/11/25, 7:00 AM

can we get z.ai added to the list of provider? i tried adding it custom and couldnt get it to work

piro•12/11/25, 2:18 PM

Does Unified Billing only work when there are no registered keys as BYOK? For example, I want to use Grok with Unified Billing, but OpenAI's key is registered with AI Gateway. In this case, requests to Gemini fail.

piro•12/11/25, 3:14 PM

Perhaps the streamGenerateContent function is not working with the combination of AI Gateway and Gemini. It worked with models like Grok, but not with Gemini.

piro•12/11/25, 3:20 PM

Regarding Gemini, after registering BYOK, it worked with streamGenerateContent.
This means it seems not to support Unified Billing. However, the documentation states that it is supported.

Ppiro Regarding Gemini, after registering BYOK, it worked with streamGenerateContent. ...

LemmoTrestoOP•12/11/25, 7:10 PM

Yeah I’ve had this issue as well and dropped it in here but had no response.

LemmoTrestoOP•12/11/25, 7:11 PM

I see google vertex was removed as a provider from the ai-gateway-provider package and the commit says it no longer works with BYOK? Why can I still provide one in dashboard then? Will this be added back or is this permanent?

R•12/12/25, 6:18 AM

i have about 1 million requests running through a gateway in 24 hours (~11 req/s), it seems like the logs are taking some time to show up in the cloudflare ai gateway dashboard. it could take anytime between 1-3 hours for it to be updated. is this normal?

Palash•12/14/25, 7:21 AM

Did Cloudflare remove Evaluations from

Palash•12/15/25, 3:11 PM

Anyone know if gemini-2.5-flash-native-

Jez•12/16/25, 12:19 AM

idk if its just me/claude but it was quite hard to get gpt oss working on the ai gateway, took a few hours of looking at lots of docs pages and such for claude to figure out how to get it working compared to having no issues at all getting qwen working for example

Sb•12/16/25, 4:49 AM

Will replicate acquisition going to merge with the ai gateway of cf ?
Because they also have similar kind of working (using same credits to run any model).

Is it so?
Or it will be a part of worker ai in which we can use neurons ?

Is it known that AI Gateway with Unified Billing (Beta) doesn’t seem to work with OpenAI when using

Similar Threads