Yes, exactly. Which is not what I'd expect. Of course it's just an impression but it's a bit weird f

Yes, exactly. Which is not what I'd expect. Of course it's just an impression but it's a bit weird for me when I call the endpoint the first time (a decent response) and a second/third time with the same parameters (now it's a much shorter and less informative response).

AAli_hat Yes, exactly. Which is not what I'd expect. Of course it's just an impression bu...

℠

℠•6/20/24, 7:24 AM

That’s the randomness at play

℠

℠•6/20/24, 7:25 AM

Maybe you want to fine tune it and train it to provide more consistent answers in the form that you need?

℠

℠•6/20/24, 7:26 AM

Asking GPT the exact same question would also yield me sometimes very different answers - don’t know if that’s changed, as I haven’t been working too much with the chat models

℠

℠•6/20/24, 7:29 AM

In SD you can provide a fixed seed, which means you can keep reloading the same prompt until the style details are some you like. Then you can lock that seed in and continue working on variants/changes but instead of entirely different outputs, it stays in the vicinity of the original image

℠

℠•6/20/24, 7:30 AM

So you can make alts with a similar looking character in the same landscape and style, etc

℠

℠•6/20/24, 7:30 AM

Maybe you can do something akin to that with yours

℠℠Maybe you want to fine tune it and train it to provide more consistent answers i...

Ali_hatOP•6/20/24, 7:31 AM

Yes, that's eventually what I'd like to do. And getting different answers is usually fine. I had hoped to squeeze out a bit more without finetuning but I'm maybe being too naive.

℠℠In SD you can provide a fixed seed, which means you can keep reloading the same ...

Ali_hatOP•6/20/24, 7:31 AM

That's a good hint, I haven't experimented with using seed yet.

℠

℠•6/20/24, 7:32 AM

As it sounds like you don’t really need a dynamic bot for long conversation, but just one that does single a Q&A at a time

Phonebox•6/21/24, 9:26 PM

Is anyone running langchain on workers? Curious to get some insights into examples of costs

Mmichelle hm sending you a dm

hydroxide•6/22/24, 3:29 AM

Sure, feel free to DM me

sarme•6/22/24, 4:26 PM

Is there any plan to support an image embedding model in the api?

vrballer•6/23/24, 12:56 AM

anyone know of an open source model better than whisper? If not anyone tried training it?

Vvrballer anyone know of an open source model better than whisper? If not anyone tried tra...

Isaac McFadyen•6/24/24, 2:34 AM

Depends. Nvidia's Parakeet is quite good at noisy audio (far better than Whisper in my experience). Haven't tried training/finetuning it but the base model is quite good.

Isaac McFadyen•6/24/24, 2:34 AM

Parakeet has two versions - one with no punctuation that is best for accuracy (but is missing punctuation) and one that has punctuation like Whisper that has a little less accuracy.

Isaac McFadyen•6/24/24, 2:35 AM

If you go with the one that doesn't have punctuation there are other models designed to take text as input and add punctuation that you can run on the text after.

German•6/24/24, 5:17 PM

What's the highest token count LLM available on workers ai?

codebam•6/25/24, 10:24 PM

I get this error after updating cloudflare-types https://p.seanbehan.ca/d337e

codebam•6/25/24, 10:24 PM

Argument of type '"@hf/thebloke/deepseek-coder-6.7b-instruct-awq"' is not assignable to parameter of type 'BaseAiImageToTextModels'. typescript (2769) [84, 37]

Argument of type '"@hf/thebloke/deepseek-coder-6.7b-instruct-awq"' is not assignable to parameter of type 'BaseAiImageToTextModels'. typescript (2769) [84, 37]

codebam•6/25/24, 10:28 PM

the code still works if I

// @ts-expect-error

// @ts-expect-error

Ccodebam I get this error after updating cloudflare-types https://p.seanbehan.ca/d337e

kian•6/25/24, 11:07 PM

What version of workers-types?

Kkian What version of workers-types?

codebam•6/26/24, 12:36 AM

4.20240620.0

GGerman What's the highest token count LLM available on workers ai?

Raylight•6/26/24, 10:17 AM

Token count as in context window? qwen1.5-7b-chat-awq can handle a context of ~20k tokens, qwen1.5-14b-chat-awq ~7.5k, and llama-3-8b-instruct-awq ~8k. (Based on my testing.) Those are also limits for the total number of tokens, including output.

Raylight•6/26/24, 12:40 PM

Hmm.. interesting..

@cf/mistral/mistral-7b-instruct-v0.2-lora

@cf/mistral/mistral-7b-instruct-v0.2-lora

has a context (and total limit) of ~15k tokens and you don't need a LoRA to run it.

云

云•6/27/24, 6:52 AM

huh i tried to pass number array as image to dreamshaper

云

云•6/27/24, 6:52 AM

but got the error

3010: Invalid or incomplete input for the model: model returned: [request id: 41fff8e6-617c-439f-a093-f02bfa2d45bb] unexpected shape for input 'image' for model 'dreamshaper-8-lcm'. Expected [1], got [1,1048265].

3010: Invalid or incomplete input for the model: model returned: [request id: 41fff8e6-617c-439f-a093-f02bfa2d45bb] unexpected shape for input 'image' for model 'dreamshaper-8-lcm'. Expected [1], got [1,1048265].

yolossn•6/27/24, 10:30 AM

Hey folks,
I am getting authentication errors when using the workers AI, the curl and python requests that is auto generated also is failing.
{'success': False, 'errors': [{'code': 10000, 'message': 'Authentication error'}]}

云云 huh i tried to pass number array as image to dreamshaper

Raylight•6/27/24, 11:48 AM

@cf/lykon/dreamshaper-8-lcm

@cf/lykon/dreamshaper-8-lcm

doesn't support passing an image. try

@cf/runwayml/stable-diffusion-v1-5-img2img

@cf/runwayml/stable-diffusion-v1-5-img2img

if you want to do image-to-image.

PPerry Sahnow Is it possible to list all models in a similar fashion to the wrangler command s...

Victor•6/28/24, 7:26 PM

API call to the cf: https://developers.cloudflare.com/api/operations/workers-ai-search-model
If you care about custom upload loras: https://developers.cloudflare.com/api/operations/workers-ai-list-finetunes
public loras: https://developers.cloudflare.com/api/operations/workers-ai-list-public-finetunes

Cloudflare API Documentation

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

Cloudflare API Documentation

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

Cloudflare API Documentation

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

Victor•6/28/24, 7:30 PM

You could also turn it into a ci build step that runs on a schedule and include it in your release
I did it here in GA in bash: https://github.com/demosjarco/matt-ai/blob/production/.github/workflows/ai-catalog.yml
End result (json): https://github.com/demosjarco/matt-ai/blob/production/shared/workers-ai-catalog.json
End result (ts): https://github.com/demosjarco/matt-ai/blob/production/shared/workers-ai-catalog.ts

GitHub

matt-ai/.github/workflows/ai-catalog.yml at production · demosjarco...

Magically All The Things AI. Contribute to demosjarco/matt-ai development by creating an account on GitHub.

GitHub

matt-ai/shared/workers-ai-catalog.json at production · demosjarco/m...

Magically All The Things AI. Contribute to demosjarco/matt-ai development by creating an account on GitHub.

GitHub

matt-ai/shared/workers-ai-catalog.ts at production · demosjarco/mat...

Magically All The Things AI. Contribute to demosjarco/matt-ai development by creating an account on GitHub.

VVictor API call to the cf: https://developers.cloudflare.com/api/operations/workers-ai-...

Victor•6/28/24, 7:32 PM

@Perry Sahnow because these count towards the global rate limit (1200/5m - https://developers.cloudflare.com/fundamentals/api/reference/limits/), you could cache the api results or go the build step method

PPerry Sahnow Yeah they presumably don't change more often than daily

Victor•6/28/24, 7:40 PM

I personally have mine set to update weekly and just watch and see how big of a commit (if any) happens in a given week - adjust from there. I did ask michelle about scheduled update cycle of models going forward, but they don't have any planned yet

云

云•6/28/24, 11:42 PM

huh i think the ai isn't recognizing my tools?

3010: Invalid or incomplete input for the model: model returned: [request id: 41fff8e6-617c-439f-a093-f02bfa2d45bb] unexpected shape for input 'image' for model 'dreamshaper-8-lcm'. Expected [1], got [1,1048265].

Yes, exactly. Which is not what I'd expect. Of course it's just an impression but it's a bit weird f

Similar Threads

Similar Threads

Similar Threads