please tell me the files supported in cf/whisper, no files work for me

Iicemelt_melt please tell me the files supported in cf/whisper, no files work for me

T

Tony Leung•9/29/24, 5:13 PM

have you tried wav format?

L

LM•9/30/24, 12:12 AM

Does anyone know how I can check if a model exists in the cloudflare API?

L

LM•9/30/24, 12:12 AM

Using the rest api

T

tadhglewis•9/30/24, 3:35 AM

Potentially dumb question (haven't used Workers AI much before) is it possible to parse (into a normalised object) and/or summarise pdfs and doc files? ChatGPT allows you to upload arbitary files and parse/summarise them. I know it's possible for images?

Eechoes221 I'm trying to get started with workers ai function calling, for the life of me I...

C

cookie•9/30/24, 9:33 AM

This would be nice, structured outputs/json mode

Ttadhglewis Potentially dumb question (haven't used Workers AI much before) is it possible t...

B

Brett (Fiberplane)•9/30/24, 3:37 PM

there is an example "chat with PDF" app that could help? https://x.com/rafalwilinski/status/1830911922805551560

Rafal Wilinski (@rafalwilinski) on X

Can you build a full-stack AI app on @cloudflare? The answer is: absolutely.

Introducing "Fullstack Cloudflare RAG" aka Chat with PDFs solution built 100% on Cloudflare using only their services.

Link to demo

Twitter

•

9/3/24, 10:13 AM

K

Kraii•9/30/24, 4:56 PM

are the beta models still free and unlimited?

K

Kraii•9/30/24, 4:57 PM

or does it count towards the free limit now

T

tadhglewis•9/30/24, 5:57 PM

https://developers.cloudflare.com/workers-ai/platform/pricing/

Cloudflare Docs

Pricing | Cloudflare Workers AI docs

Workers AI is included in both the Free and Paid Workers plans and is priced based on model task, model size, and units.

K

Kraii•9/30/24, 6:16 PM

yeah I checked that, it doesn't mention about beta models so I came here to confirm

T

tadhglewis•9/30/24, 6:26 PM

Standard models are large image models such as @cf/stabilityai/stable-diffusion-xl-base-1.0...

It lists this model in the pricing, which is beta.

C

cloudwhere?•9/30/24, 8:54 PM

Hi everyone! I need some help with integrating Cloudflare with mailchimp. Specifically, I’m looking for guidance on how to copy and paste the CNAME and A Record codes from mailchimp into Cloudflare in the right places without messsing up my company's connections, Where exactly should I put them? Any tips or steps would be greatly appreciated

T

thousandmiles•10/1/24, 2:57 AM

does gemini on workers ai support streaming and tool use?

R

Razmjoo•10/2/24, 1:20 PM

Hi everyone,

is there any way that we can request a new embedding model to be added to the list? one of the top MTEB Leaderboard > Bitext models is not available, and it would be great to have it in deployed models for workers

https://huggingface.co/intfloat/multilingual-e5-large

intfloat/multilingual-e5-large · Hugging Face

R

Rakhim•10/2/24, 1:57 PM

Hi everyone,
I'm new to LLMs, and I'm wondering what could I use to classify text into tags (I want to auto-tag blog posts, for example).

@cf/huggingface/distilbert-sst-2-int8

@cf/huggingface/distilbert-sst-2-int8

@cf/huggingface/distilbert-sst-2-int8

@cf/huggingface/distilbert-sst-2-int8 can seemingly only do sentiment classification.

�

😈 Donkey 💫•10/2/24, 7:29 PM

const validModel = ['gpt-4o-mini', 'gpt-4o-mini-2024-07-18', 'gpt-4o', 'gpt-4o-2024-08-06', 'gpt-3.5-turbo-0125', 'gpt-3.5-turbo-1106']; // NO o1-mini
export default class extends WorkerEntrypoint<Env> {
    async fetch(request: Request): Promise<Response> {
        const { model, messages }: any = await request.json();
        return this.rpc(model, messages);
    }

    async rpc(_model: string, messages: Messages[], colo?: string): Promise<Response> {
        const model = _model?.split('/')[1]; // slice prefix openai/
        if (!model || !validModel.includes(model)) {
            return new Response(JSON.stringify({ message: 'Invalid model' }), { status: 400 });
        }

...
        const aiPayload = {
            model: model,
            messages: messages,
            max_tokens: 2048,
            stream: true
        }
...
        const response = await fetch(cfGateway, {
            method: 'POST',
            headers: {
                'Content-Type': 'application/json',
                'Authorization': `Bearer ${token}`
            },
            body: JSON.stringify(aiPayload),

        });

const validModel = ['gpt-4o-mini', 'gpt-4o-mini-2024-07-18', 'gpt-4o', 'gpt-4o-2024-08-06', 'gpt-3.5-turbo-0125', 'gpt-3.5-turbo-1106']; // NO o1-mini
export default class extends WorkerEntrypoint<Env> {
    async fetch(request: Request): Promise<Response> {
        const { model, messages }: any = await request.json();
        return this.rpc(model, messages);
    }

    async rpc(_model: string, messages: Messages[], colo?: string): Promise<Response> {
        const model = _model?.split('/')[1]; // slice prefix openai/
        if (!model || !validModel.includes(model)) {
            return new Response(JSON.stringify({ message: 'Invalid model' }), { status: 400 });
        }

...
        const aiPayload = {
            model: model,
            messages: messages,
            max_tokens: 2048,
            stream: true
        }
...
        const response = await fetch(cfGateway, {
            method: 'POST',
            headers: {
                'Content-Type': 'application/json',
                'Authorization': `Bearer ${token}`
            },
            body: JSON.stringify(aiPayload),

        });

const validModel = ['gpt-4o-mini', 'gpt-4o-mini-2024-07-18', 'gpt-4o', 'gpt-4o-2024-08-06', 'gpt-3.5-turbo-0125', 'gpt-3.5-turbo-1106']; // NO o1-mini
export default class extends WorkerEntrypoint<Env> {
    async fetch(request: Request): Promise<Response> {
        const { model, messages }: any = await request.json();
        return this.rpc(model, messages);
    }

    async rpc(_model: string, messages: Messages[], colo?: string): Promise<Response> {
        const model = _model?.split('/')[1]; // slice prefix openai/
        if (!model || !validModel.includes(model)) {
            return new Response(JSON.stringify({ message: 'Invalid model' }), { status: 400 });
        }

...
        const aiPayload = {
            model: model,
            messages: messages,
            max_tokens: 2048,
            stream: true
        }
...
        const response = await fetch(cfGateway, {
            method: 'POST',
            headers: {
                'Content-Type': 'application/json',
                'Authorization': `Bearer ${token}`
            },
            body: JSON.stringify(aiPayload),

        });

const validModel = ['gpt-4o-mini', 'gpt-4o-mini-2024-07-18', 'gpt-4o', 'gpt-4o-2024-08-06', 'gpt-3.5-turbo-0125', 'gpt-3.5-turbo-1106']; // NO o1-mini
export default class extends WorkerEntrypoint<Env> {
    async fetch(request: Request): Promise<Response> {
        const { model, messages }: any = await request.json();
        return this.rpc(model, messages);
    }

    async rpc(_model: string, messages: Messages[], colo?: string): Promise<Response> {
        const model = _model?.split('/')[1]; // slice prefix openai/
        if (!model || !validModel.includes(model)) {
            return new Response(JSON.stringify({ message: 'Invalid model' }), { status: 400 });
        }

...
        const aiPayload = {
            model: model,
            messages: messages,
            max_tokens: 2048,
            stream: true
        }
...
        const response = await fetch(cfGateway, {
            method: 'POST',
            headers: {
                'Content-Type': 'application/json',
                'Authorization': `Bearer ${token}`
            },
            body: JSON.stringify(aiPayload),

        });

Someone from Russia was trying to break my backend worker with model

o1-mini

o1-mini

o1-mini

o1-mini
My worker is private without public endpoint, which is invoked by service binding via rpc.

I can't understand how this happens.
How o1-mini can bypass my validModel check.

The image is shot from logs AI gateway dashboard.

Any potential vulnerable here?

�😈 Donkey 💫```ts const validModel = ['gpt-4o-mini', 'gpt-4o-mini-2024-07-18', 'gpt-4o', 'gp...

I

Isaac McFadyen•10/2/24, 8:56 PM

Have you double-checked whether they were actually hitting the Worker or were hitting AI Gateway directly? I suspect that OpenAI returns "unsupported country" before checking the authentication, so if they found your AI Gateway URL and were hitting that then you'd see that.

�

😈 Donkey 💫•10/3/24, 3:38 AM

he was hitting my domain worker, 90% for sure

payload scheme was pre-defined same as inside my worker.
Analytics logs have corresponding requests at the same time too

�😈 Donkey 💫he was hitting my domain worker, 90% for sure payload scheme was pre-defined sa...

F

forbiele•10/3/24, 3:47 AM

W hit ngl

�

😈 Donkey 💫•10/3/24, 4:11 AM

i think i found out issue, my teammate was adding o1-mini for testing for a while then removed it.
that's the same time Rus hacker trying to bypass

R

Robert•10/3/24, 8:21 AM

Hi, I'm using openai python library with cloudflare ai as base. I can pass text content to it clearly but I can't figure out how to send images to it. I've seen the tutorial "floor-llava" but it's for workers-typescript not openai-python. I've even checked openai platform but it's not supporting in cloudflare. I'm using "@cf/meta-llama-3.2-11b-vision-instruct" Anyone can help this issue?
Thanks in advance.

A

Ashley•10/3/24, 6:09 PM

I'm struggling to see the usage of Workers AI, I am using

@cf/baai/bge-large-en-v1.5

@cf/baai/bge-large-en-v1.5

@cf/baai/bge-large-en-v1.5

@cf/baai/bge-large-en-v1.5 but when I go to the Workers AI page in the dash, I just have the "Welcome" page

U

unsigned•10/6/24, 1:11 AM

response

R

Ray Li•10/8/24, 4:36 AM

does any one know how to create workers ai resources using pulumi?

R

Ray Li•10/8/24, 4:36 AM

Couldn't find it on the docs, seems like we have to use wrangler?

K

Keebs•10/8/24, 8:28 PM

RRay Li does any one know how to create workers ai resources using pulumi?

K

Keebs•10/8/24, 8:29 PM

https://developers.cloudflare.com/pulumi/tutorial/hello-world/

A

Atinux•10/9/24, 1:37 PM

I made an open source template to generate image using Flux-1 Schnell & R2 to store generated images
https://github.com/atinux/flux-ai-image-generator

GitHub

GitHub - atinux/flux-ai-image-generator: Generate images with AI us...

Generate images with AI using Flux-1 Schnell running on your Cloudflare account. - atinux/flux-ai-image-generator

A

Atinux•10/9/24, 1:37 PM

I have a question regarding the options, is it possible to specify the

width

width

width

width &

height

height

height

height of the image?

A

Atinux•10/9/24, 1:38 PM

I know that Together AI can support this but I rather use Cloudflare and wanted to know if this will be possible

AAtinux I made an open source template to generate image using Flux-1 Schnell & R2 to st...

A

Advany•10/9/24, 1:38 PM

No they removed all options

AAdvany No they removed all options

A

Atinux•10/9/24, 1:38 PM

Do you know why?

AAtinux Do you know why?

A

Advany•10/9/24, 1:51 PM

No. I just saw a update github for the docs

A

Advany•10/10/24, 7:29 PM

Finally got the hang of prompting flux schnell. Really loving it!

P

Phanindra Reddy•10/11/24, 2:39 AM

Hey guys, when I am trying to access the llama 3.2 model, it is showing me an error of model not found, is it only for me or the model is working?

AAdvany Finally got the hang of prompting flux schnell. Really loving it!

P

Phanindra Reddy•10/11/24, 2:39 AM

Any tips?? I just made a image generator for it

Oorangebrain Hi everyone, I'm currently working with the Cloudflare-hosted BAAI model (@cf/ba...

R

Rihan•10/11/24, 4:32 AM

I'm experiencing the same issue. Did you manage to find the cause?

RRobert Hi, I'm using openai python library with cloudflare ai as base. I can pass text ...

C

Craig Dennis•10/11/24, 4:38 PM

Just seeing this, do you mean you are using the openai_compat?

PPhanindra Reddy Any tips?? I just made a image generator for it

A

Advany•10/11/24, 6:58 PM

prompt the style you like to use

J

Julius•10/13/24, 4:02 PM

hi guys, i want to ask something about workers ai, i created a Hono app to be deployed on cloudflare worker

I already done to binds the AI in the wrangler.toml, but as i want to access env.AI.run its say

X [ERROR] TypeError: Cannot read properties of undefined (reading 'run')

X [ERROR] TypeError: Cannot read properties of undefined (reading 'run')

X [ERROR] TypeError: Cannot read properties of undefined (reading 'run')

X [ERROR] TypeError: Cannot read properties of undefined (reading 'run')

Can anyone help me?

JJulius hi guys, i want to ask something about workers ai, i created a Hono app to be de...

I

Isaac McFadyen•10/13/24, 4:21 PM

In Hono,

env

env

is on

c

I

Isaac McFadyen•10/13/24, 4:21 PM

So it would be

c.env.AI

c.env.AI

c.env.AI

c.env.AI

IIsaac McFadyen So it would be `c.env.AI`

J

Julius•10/13/24, 4:29 PM

Wow! Thanks a lot for your help! I was new in Hono, i'll be exploring it more :plus1green:

I

Isaac McFadyen•10/13/24, 4:29 PM

No problem

P

PRM•10/13/24, 6:46 PM

Hey i am making my portfolio and in that i want ai to give information about the title, like it should generate text in the card by reading the title of the card so is it free on cloudfare?

F

Flare•10/13/24, 6:49 PM

Please do not post your question in multiple channels/post it multiple times per the rules at ⁠#

welcome-and-rules. It creates confusion for people trying to help you and doesn't get your issue or question solved any faster.

P

PRM•10/13/24, 6:49 PM

ohkk sry

please tell me the files supported in cf/whisper, no files work for me

Similar Threads

Similar Threads

Similar Threads