Hey! our AI requests are failing with the following rate limit error, while the volume is relatively

Hey! our AI requests are failing with the following rate limit error, while the volume is relatively low ($20 of consumption). That heavily impacts our services

 {       message: 'AiError: AiError: Capacity temporarily exceeded, please try again.',       code: 3040     }

 {       message: 'AiError: AiError: Capacity temporarily exceeded, please try again.',       code: 3040     }

LLila Rest Hey! our AI requests are failing with the following rate limit error, while the ...

blindChicken•2/13/25, 11:28 PM

I was getting the same error, but according to my dashboard I was no where close to the my limit. I think the error is not specific to any one account but may be that the network is overloaded. (I'm speculating) I waited a few minutes and I was able to continue normally.

TTanis Wolfbane Any one manage to get flux to work? When I run the exactly example in the docume...

vehu•2/14/25, 1:57 AM

hey, have you found any solution to the problem? (sorry for the ping)

BblindChicken I think the error may be caused because you did not provide a description for yo...

Jason Barry•2/14/25, 2:53 AM

do you know if this requirement changed recently? it seems that last week, there wasn't enforcement of the

description

description

key

JJason Barry do you know if this requirement changed recently? it seems that last week, there...

blindChicken•2/14/25, 3:29 AM

IDK about enforcement, but it is needed so that the model can assign the proper value to the proper properties. Since all this AI stuff is probabilistic it is possible that it can work without the description, and then just as easily not work.

Bbeingmudit [WePlayDOS]CF team, can anyone help with yhe above?

�

😈 Donkey 💫•2/14/25, 5:00 AM

only 1 ancient model on CF support Func Calling, even when your payload works, it would return something poor-quality and irrelevant, that's my experience

LLila Rest Hey! our AI requests are failing with the following rate limit error, while the ...

�

😈 Donkey 💫•2/14/25, 5:00 AM

since CF list their inference on OR, I encounter the same problem

BblindChicken I was getting the same error, but according to my dashboard I was no where close...

Lila RestOP•2/14/25, 9:23 AM

Same it finally solved by itself after a few hours. That sucks :/

LLila Rest Same it finally solved by itself after a few hours. That sucks :/

blindChicken•2/14/25, 10:08 PM

I agree it sucks, and it would be nice if someone from Cloudflare could chime in a clarify exactly what the error really is:

{ message: 'AiError: AiError: Capacity temporarily exceeded, please try again.', code: 3040 }

{ message: 'AiError: AiError: Capacity temporarily exceeded, please try again.', code: 3040 }

atgctg•2/17/25, 7:20 AM

Tried image generation today, some feedback:

would need 9:16 aspect ratio, only 1:1 is supported

type issues (

@cf/black-forest-labs/flux-1-schnell

@cf/black-forest-labs/flux-1-schnell

is not available as a

BaseAiTextToImageModels

BaseAiTextToImageModels

)

for
```
schnell
```
```
schnell
```
, p90 latency should be ~500 ms, not seconds
bindings based api is nice!

Defy•2/17/25, 2:46 PM

Sry is i missunderstand, but i am RN on the free plan, using my 10.000 AI Neurons a day and ran into a problem

The LLM response seems to be arbitrarily cut off at 500 tokens.
is there such a policy? is there a way to disable this or is it just a code issue on my part?

LLiam Error processing request: 3040: Capacity temporarily exceeded, please try again....

Defy•2/17/25, 3:39 PM

Been getting similar errors when trying to get over 10 requests / second

LLiam The error is an issue CF end tho not the Rate Limit so its a bit scuffed really....

Defy•2/17/25, 3:40 PM

ah.
i can't help you then.

btw,
any clue how to get more than 500 tokens out of the free tier llm?

blindChicken•2/17/25, 9:48 PM

Has anyone tried tool calling with

Deepseek-R1

Deepseek-R1

? When I tried it,the model simply returns a normal response as if tools where not specified. I used the schema that works

Hermes

Hermes

and

LLama

LLama

, and I have also tried the schema shown here: https://api-docs.deepseek.com/guides/function_calling. But neither work.

ʙʏᴋᴀɢᴀɴ•2/18/25, 2:33 PM

help emergency pls (cloud flare)

ʙʙʏᴋᴀɢᴀɴ help emergency pls (cloud flare)

MattIPv4•2/18/25, 2:58 PM

You're gonna want to give a bunch more context if you're expecting folks to help

ʙʙʏᴋᴀɢᴀɴ help emergency pls (cloud flare)

Wraith•2/18/25, 3:20 PM

call 911 now bass drop

KKarl Hi everyone, i'm trying to get function calling working on workers ai, but for t...

blindChicken•2/19/25, 2:36 AM

What models have you tried?

blindChicken•2/19/25, 2:42 AM

This tutorial provides a good overview, I've been able to get it to work with LLama3LLama3 and Hermes MistralHermes Mistral. Hermes is used in the tutorial.
https://www.youtube.com/watch?v=Id5oKCa__IA
I have not be able to get it to work with deepseek-r1deepseek-r1

YouTubeCloudflare Developers

Tool Calling Also Known as Function Calling on Cloudflare Workers AI

Tool calling, also known as function calling, is a powerful concept that lets you build Large Language Model based applications that can perform actions and retrieve external information from defined tools.

Repo: https://github.com/craigsdennis/lightbulb-moment-tool-calling

Sign up for Cloudflare for free!!
https://workers.cloudflare.com/

#cl...

Giovanni Cocco•2/19/25, 1:02 PM

Hey everyone! is anyone using Worker AI with OpenWeb UI?

KKarl @blindChicken ; i got it _mostly_ working with llama 3.1 still does some funky ...

blindChicken•2/19/25, 2:46 PM

For tool calling I find that

Hermes

Hermes

performs best. It produce reliable results, but even it can at times provide a messed up response (but it is rare).

BBumblebeeSquare Empty message

Raylight•2/21/25, 10:19 PM

As a follow-up, support for usage stats have now been added to the workers-ai-provider package.

Jumaron•2/22/25, 3:44 PM

Hello, would it be possible that CloudFlare adds Zonos from Zyphra for a Voice Model? It is open Source under the Apache License, so it should fit perfectly to Cloudflare.

ANBU•2/23/25, 6:20 PM

Hey, anyone faced this issue where worker ai stops responding in mid of sentence?

AANBU Hey, anyone faced this issue where worker ai stops responding in mid of sentence...

LeakyAssassin•2/23/25, 6:22 PM

it's fake ai. :D. what do you expect. Usually its from overload of tokens though.

AANBU Hey, anyone faced this issue where worker ai stops responding in mid of sentence...

kian•2/23/25, 7:25 PM

max_tokens defaults to 256, you’re probably hitting that many output tokens so it’ll stop

LLeakyAssassin it's fake ai. :D. what do you expect. Usually its from overload of tokens thou...

ANBU•2/23/25, 7:48 PM

Yeah with this thinking models like deepseek are basically unusable

Kkian max_tokens defaults to 256, you’re probably hitting that many output tokens so i...

ANBU•2/23/25, 8:21 PM

Is there a way to increase this limit?

ANBU•2/23/25, 9:49 PM

My bad, missed that in the docs

ANBU•2/23/25, 9:49 PM

Thanks!

Ben-xD•2/24/25, 9:58 AM

Does it make sense to a workers AI image-to-text use for image content moderation? Or should i avoid workers AI for that

mr.niko.la•2/25/25, 4:29 AM

Faster Latency sorry not sure if that is faster response or slower response times. I assume you mean faster.

Maybe I should run a container on my server when I can. Justify the cost to run a server

Mmr.niko.la Faster Latency sorry not sure if that is faster response or slower response time...

Isaac McFadyen•2/25/25, 4:43 AM

Latency of Groq is almost certainly going to be faster in this case.

Florencia•2/25/25, 5:29 AM

@Isaac McFadyen

Original message was deleted

DragoDiKomo•2/25/25, 1:49 PM

I tried too to set manually the max context length but still cutting the responses

jLynx•2/25/25, 7:37 PM

We have thousands of users of our service complaining because the users LLM requests are being blocked when trying to use Anthropic from a supported country using the AI gateway through a CF Worker, but the issue is they are being routed through a HK server (Which is not supported by anthropic), so the users end up getting blocked. What can we do here to fix this critical issue @Walshy | Workers/Pages ?
We are in talks with Anthropic about this too

JjLynx We have thousands of users of our service complaining because the users LLM requ...

Isaac McFadyen•2/25/25, 8:02 PM

Please don't ping employees, Walshy isn't on the AI team so wouldn't be able to resolve this anyway.

IIsaac McFadyen Please don't ping employees, Walshy isn't on the AI team so wouldn't be able to ...

jLynx•2/25/25, 8:03 PM

Sorry, its related to workers/pages too

Isaac McFadyen•2/25/25, 8:04 PM

Walshy specifically works on the Worker deployment team so isn't able to help here. I've sent this to someone on the AI Gateway team though.

Hey! our AI requests are failing with the following rate limit error, while the volume is relatively

Similar Threads

Similar Threads

Similar Threads