Hey, looks like you are using a gateway, how can I change my code to also use a gateway instead the

Hey, looks like you are using a gateway, how can I change my code to also use a gateway instead the direct reqeust?

const embeddings: any = await env.AI.run("@cf/baai/bge-m3", {
text: [text],
});

fortytwo•9/18/25, 6:15 AM

Can a folder within a bucket be used as a RAG source? Or do I need to create separate buckets for independent rag data?

VVikash Hey, looks like you are using a gateway, how can I change my code to also use a ...

.r20•9/18/25, 6:22 AM

https://developers.cloudflare.com/ai-gateway/usage/providers/workersai/

Cloudflare Docs

Workers AI

Use AI Gateway for analytics, caching, and security on requests to Workers AI. Workers AI integrates seamlessly with AI Gateway, allowing you to execute AI inference via API requests or through an environment binding for Workers scripts. The binding simplifies the process by routing requests through your AI Gateway with minimal setup.

Lami•9/18/25, 10:29 AM

@cf/google/embeddinggemma-300m

@cf/google/embeddinggemma-300m

Does anyone know the price of this model?

VVikash Getting too many errors: AiError: 3040: Capacity temporarily exceeded, please t...

samjs•9/18/25, 3:26 PM

Hey @Vikash and others. Apologies for the 3040 errors you all are seeing. We're aware of these issues (we get alerted when they happen) and we're working to (a) provision more capacity for those models, and (b) implement some additional measures to keep capacity available even when we're getting flooded with requests. Appreciate your patience and thanks for flagging these issues!

Ffortytwo Can a folder within a bucket be used as a RAG source? Or do I need to create sep...

samjs•9/18/25, 6:15 PM

I believe we only support using the entire bucket. #autorag channel might know more about this however!

Ssamjs I believe we only support using the entire bucket. #autorag channel might know ...

fortytwo•9/18/25, 6:23 PM

Ah I missed that space. Thanks!

Razmjoo•9/18/25, 6:36 PM

Hi, is there any outages with AI Workers? I am getting a lot of errors on text-embeddings workers (using bge-m3) cc: @samjs

{
  "errors": [
    {
      "message": "AiError: AiError: model temporarily unavailable (58da3f75-5ff1-46ba-a45e-a59d4ada1c33)",
      "code": 9000
    }
  ],
  "success": false,
  "result": {},
  "messages": []
}

{
  "errors": [
    {
      "message": "AiError: AiError: model temporarily unavailable (58da3f75-5ff1-46ba-a45e-a59d4ada1c33)",
      "code": 9000
    }
  ],
  "success": false,
  "result": {},
  "messages": []
}

Ssamjs Hey @Vikash and others. Apologies for the 3040 errors you all are seeing. We're ...

Isaac McFadyen•9/18/25, 6:38 PM

@Razmjoo see this message

IIsaac McFadyen @Razmjoo see this message

Razmjoo•9/18/25, 6:41 PM

Are there any timeline e.g, do we expect it in coming hours to get fixed; or is it that might go over weekend/next week... just thinking if we need a fallback plan for business continuity ...

engnadeau•9/18/25, 7:08 PM

i thought i was going crazy on my end. is there not a status page for Workers AI?

engnadeau•9/18/25, 7:08 PM

AiError: 9000: model temporarily unavailable

AiError: 9000: model temporarily unavailable

when trying to use

@cf/baai/bge-m3

@cf/baai/bge-m3

Eengnadeau `AiError: 9000: model temporarily unavailable` when trying to use `@cf/baai/bge-...

ajgeiss0702•9/18/25, 7:12 PM

im seeing this too

engnadeau•9/18/25, 7:24 PM

@samjs is the 9000 error related to the 3040 errors? is there a status page for Workers AI and model capacity? https://www.cloudflarestatus.com/ says all systems operational

Cloudflare Status

Welcome to Cloudflare's home for real-time and historical data on system performance.

samjs•9/18/25, 7:27 PM

The 9000 and 3040 errors are related -- we're seeing a huge influx of requests. Deploying some config changes now to try and help with that.

The follow up work I mentioned before should help us deal with this situations better in the future.

is there a status page for Workers AI and model capacity? https://www.cloudflarestatus.com/ says all systems operational

Currently the status page reflects the global status, not specific models.

James•9/18/25, 7:39 PM

Specific models being inoperable could result in 100% unavailability for some people's applications. I'd encourage y'all to reconsider your approach to status reporting here in future please so that these things can be better communicated to folks who might not just be in the Discord. :MeowHeartCloudflare:

samjs•9/18/25, 7:53 PM

https://www.cloudflarestatus.com/incidents/nrxwc9tny0z7

Workers AI Errors for bge-m3 Model

�

𝗞enneth•9/18/25, 8:16 PM

Hi Cloudflare team,

I've been working with AutoRAG and it's a great feature for building RAG systems. However, I notice it's currently limited to models available in Workers AI.
Is there any plan to support custom model providers and custom embedding providers in AutoRAG?

Specifically:
Custom LLM providers (OpenAI, Claude, ETC)
Custom embedding providers (OpenAI embeddings, specialized models)
Bring-your-own-model support

This would enable using specialized embeddings for specific domains, cost optimization across providers, and better architectural flexibility for enterprise use cases.
Is this being considered for the roadmap, or is there currently an undocumented way to achieve this?

Thanks for the great work on Workers AI.

�𝗞enneth Hi Cloudflare team, I've been working with AutoRAG and it's a great feature fo...

Chaika•9/18/25, 8:39 PM

I would ask in #autorag , it's a bit weird to ask in the workers-ai channel about not using workers-ai lol

CChaika I would ask in #autorag , it's a bit weird to ask in the workers-ai channel abou...

samjs•9/18/25, 9:00 PM

Haha, that's okay we have thick skin

Does sound like a reasonable use case, but indeed you'll get a better response in the #autorag channel!

�

𝗞enneth•9/18/25, 9:20 PM

@samjs Sorry for posting in the wrong channel! I meant to ask this in #autorag. Thanks for pointing that out.

achesui•9/18/25, 11:38 PM

Hello, I'm not being able to make the new model

nova-3

nova-3

work with language spanish websocket via REST API, i'm testing this with the repository: https://github.com/cloudflare/realtime-examples/tree/main/ai-tts-stt when setting the language to 'es' or 'es-419' it gives me a 500 error: Failed to establish Nova WebSocket: 500 while the websocket keeps running, it works for english only, can anyone help me?

GitHub

realtime-examples/ai-tts-stt at main · cloudflare/realtime-examples

Contribute to cloudflare/realtime-examples development by creating an account on GitHub.

Ffortytwo Ah I missed that space. Thanks!

.r20•9/19/25, 7:56 PM

yeah itll just index the entire bucket not just the folder

..r20 yeah itll just index the entire bucket not just the folder

fortytwo•9/19/25, 8:11 PM

I did notice you can filter on directories. But not sure when multiple rag source buckets vs one bucket + multiple dirs + dir filters would be best

.r20•9/19/25, 8:40 PM

idt it matters id just do the directory that’s easiest

.r20•9/19/25, 8:40 PM

but i think autorag has some fundamental issues they need to fix so i would rather use vectorize directly its better

fortytwo•9/19/25, 8:46 PM

What issues have you seen there?

engnadeau•9/19/25, 8:51 PM

error of the day:

  error: '4002: could not route request to AI model',

  error: '4002: could not route request to AI model',

with

  model: '@cf/baai/bge-m3',

  model: '@cf/baai/bge-m3',

is this also due to high demand?

engnadeau•9/19/25, 9:07 PM

is Workers AI considered production ready?

James•9/19/25, 9:42 PM

https://www.cloudflarestatus.com/incidents/k9wl5k2hb13m

Workers AI experiencing degraded availability in some models

Ffortytwo What issues have you seen there?

.r20•9/19/25, 9:48 PM

workers ai is still a few iterations away from production ready i cant launch it to my customers in the state its in rn (although they are working on it p fast), and indexing isnt super reliable rn imo (partially due to workers ai and then other generic errors it throws)

.r20•9/19/25, 9:49 PM

for prototyping its fine but vectorize is very good and i still recommend that and not a lot more work than autorag

fortytwo•9/19/25, 9:49 PM

Yeah I did have a bunch of indexing errors when I pushed stuff, had to redo it 5 or 6 times to get all of them done

.r20•9/19/25, 9:49 PM

yeah the problem is my customers update the rag extremely frequently so i cant rely on their indexing, but like i said its still good for prototyping

fortytwo•9/19/25, 9:55 PM

Ok, well that's good to know. So you're using vectorize "manually", basically? Still storing data in a cloudflare bucket but generating and searching the vectors yourself rather than the autorag apis?

.r20•9/19/25, 10:38 PM

i store data in d1 and yeah i just make the embeddings through openai and insert into the vectorize

.r20•9/19/25, 10:38 PM

and then searching isnt that hard either

.r20•9/19/25, 10:39 PM

u can vibe code most of this stuff anyway through cloudflare's mcp server and whatever ai powered ide u use

ray•9/21/25, 4:23 AM

Is any one else getting issues with tool calls? I can only seem to get proper tool call responses via the ai sdk when I use mistral-small-3.1-24b-instruct

ray•9/21/25, 4:24 AM

All the other tool calling models either return a text type with tool params json stringified or error out

Ashkan•9/21/25, 9:53 AM

Hello, when I want to poll batch request I encounter this error:

Error: 3030: 1 validation error for VllmBatchRequest

Error: 3030: 1 validation error for VllmBatchRequest

does anyone know what doesn't mean and how to resolve it?

Ashkan•9/21/25, 9:56 AM

Gateway's logs looks hanging. it took tens of minutes but no change:

hellc•9/22/25, 1:21 PM

seems like bge-m3 over workers-ai is lagging again. Same issue from friday's evening...

.venv/lib/python3.10/site-packages/llama_index/embeddings/cloudflare_workersai/base.py", line 118, in _aget_text_embeddings
    return resp["result"]["data"]
KeyError: 'data'

.venv/lib/python3.10/site-packages/llama_index/embeddings/cloudflare_workersai/base.py", line 118, in _aget_text_embeddings
    return resp["result"]["data"]
KeyError: 'data'

Leo•9/22/25, 3:34 PM

Do you have an automated article-writing bot for your website?

blindChicken•9/22/25, 4:28 PM

I'm having an issue with workersAI, it is returning all kinds of strange errors with little information of what the problem may be. I have gotten:

InferenceUpstreamError: error code: 1031

InferenceUpstreamError: error code: 1031

and now I'm getting

InferenceUpstreamError: <!DOCTYPE html>...

InferenceUpstreamError: <!DOCTYPE html>...

where it is returning CF html error page, with 500 error. I've tried using my production worker, and then in dev with --remote and dev locally. In all cases it returns some variation of the above but all fail. The code worked fine yesterday and hasn't changed. It fails at the call

"@cf/qwen/qwen2.5-coder-32b-instruct"

"@cf/qwen/qwen2.5-coder-32b-instruct"

. All other requests appear to work.

BblindChicken I'm having an issue with workersAI, it is returning all kinds of strange errors ...

blindChicken•9/22/25, 6:02 PM

It appears that this error is being caused by having

remote = true

remote = true

included in my

wrangler.toml

wrangler.toml

I had it set in both

queues

queues

and

r2_buckets

r2_buckets

, but oddly not in

ai

ai

. Removing them fixed the problem.

BblindChicken It appears that this error is being caused by having `remote = true` included in...

blindChicken•9/22/25, 6:16 PM

Removing

remote = true

remote = true

solved the problem in dev but not in production. I redeployed worker, but the request still hangs. Nothing in logs...

fortytwo•9/23/25, 5:14 AM

@.r20 regarding your moving off of autorag...is it only literally autorag that is still in beta? Did you try using all cloudflare services but doing it more "manually", did that work ok? Or have to use thirdparty stuff to get something prod-ready?

Ffortytwo @.r20 regarding your moving off of autorag...is it only literally autorag that ...

.r20•9/23/25, 5:15 AM

launched vectorize to prod today it works great

.r20•9/23/25, 5:16 AM

its pretty easy to use

..r20 launched vectorize to prod today it works great

fortytwo•9/23/25, 5:16 AM

so are you just manually chunking, using cloudflare to generate vectors, inserting them, then searching and using one of cloudflare's llm models for the overall query?

{ "errors": [ { "message": "AiError: AiError: model temporarily unavailable (58da3f75-5ff1-46ba-a45e-a59d4ada1c33)", "code": 9000 } ], "success": false, "result": {}, "messages": [] }

Hey, looks like you are using a gateway, how can I change my code to also use a gateway instead the

Similar Threads

Similar Threads

Similar Threads