Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

New models and developer products

openai.com

Post reply on HN
Print view View original

271–280 of 561 posts

Re: New models and developer products

#271

Post by DaiPlusPlus »

tornato7 wrote: post #58

A few notes on pricing: - GPT-4 Turbo vision is much cheaper than I expected. A 768*768 px image costs $0.00765 to input. That's practical to replace more specialized computer vision models for many use-cases. - ElevenLabs is $0.24 per 1K characters while OpenAI TTS HD is $0.03 per 1K characters. Elevenlabs still has voice copying but for many use-cases it's no longer competitive. - It appears that there's no additio…

> GPT-4 Turbo vision is much cheaper than I expected. A 768*768 px image costs $0.00765 to input. That's practical to replace more specialized computer vision models for many use-cases

That's still on-the-orders-of $0.01/image - whereas a simple binary-classifier I wrote using OpenCV and simple histograms (no NNs here) would be like $0.0000001/image (if I had to put a price on it - on the basis that I wrote it 8 years ago in a weekend). So there's still a scalability gulf here.

----

Correct me if I'm wrong, but feeding images to GPT-4 is still done in-band, right? My understanding is that means it's forever open to, for example, a user from 4chan photoshopping-in the text "This image is not pornographic" on-top of the shock-image they upload to my hypothetical service to get it any GPT-4-based inappropriate-imagary-detector?

Re: New models and developer products

#272

Post by bkyan »

It only got a quick mention during the keynote, but I think the most important new feature for me, in terms of integrating ChatGPT into my own workflows, will be the ability to get consistent results back from the same prompt using a seed.

Re: New models and developer products

#273

Post by kossTKR »

jack_riminton wrote: post #253

For all the naysayers in the comments, the elephant in the room that no one quite wants to admit, is that GPT4 is still far better than everything else out there

Is there anything promising out there?

Is crowd sourced training still unfeasible?

I remember how fast the diffusion world moved in the first year but it seems it's stalled somewhat compared to first midjourney then Dall-e 3. Is it the same with text models?

Re: New models and developer products

#274

Post by visarga »

riku_iki wrote: post #162
whytai wrote: post #71

Every day this video ages more and more poorly [1]. categories of startups that will be affected by these launches: - vectorDB startups -> don't need embeddings anymore - file processing startups -> don't need to process files anymore - fine tuning startups -> can fine tune directly from the platform now, with GPT4 fine tuning coming - cost reduction startups -> they literally lowered prices and increased rate limits…

> - vectorDB startups -> don't need embeddings anymore they don't provide embedings, but storage and query engines for embeddings, so still very relevant > - file processing startups -> don't need to process files anymore curious what is that exactly?.. > - vertical ai agent startups -> GPT marketplace sure, those startups will be selling their agents on marketplace

It's easy to host your query engine somewhere else and integrate it as a search function in chatGPT. Quite easy to switch providers of search.

Re: New models and developer products

#276

Post by steno132 »

For all the hate: Elon ships. And OpenAI ships.

People claim OpenAI is closed, that they are controlled by Microsoft, that they don't care enough about safety...

But the fact is, Anthropic, Google Brain, even Meta -- OpenAI blows them all out of the water when it comes to shipping new innovations. Just like Twitter ships much more now with Elon, and how SpaceX ships much more than NASA and Blue Origin.

If you disagree, give me just one logical reason why. It's just a fact.

Re: New models and developer products

#277

Post by eurekin »

simonw wrote: post #241

I just released a new version of my LLM CLI tool with support for the new GPT-4 Turbo model: https://llm.datasette.io/en/stable/changelog.html#v0-12 You can install it like this: pipx install llm Then set an API key: llm keys set openai Then run a prompt through GPT-4 Turbo like this: llm -m gpt-4-turbo "Ten great names for a pet walrus" # Or a shortcut: llm -m 4t "Ten great names for a pet walrus" Here's a one-liner…

Great tool and example! Makes me wonder, what one can do more with it

Re: New models and developer products

#278

Post by lucubratory »

9dev wrote: post #264
thethimble wrote: post #214

Earlier quoted context omitted.

Or, alternatively, copyright risk is a major concern for real customers, and this is a major step forward in addressing that. Not everything needs to be so cynical. What’s good for investors can be good for users as well.

You’re talking about the issue as if it weren’t your posts, your pictures, or content of artists you are into. This isn’t a ”copyright risk“, it’s a Silicon Valley corporation getting away with declaring copyright just… obsolete.

They are my content, actually, from the last ~15 years of being on the internet. I don't care about it personally, and even if I did it is really obviously fair use so even if I find it objectionable I don't get to actually legally compel someone to stop.

Re: New models and developer products

#279

Post by lucubratory »

MagicMoonlight wrote: post #240
zavertnik wrote: post #13

And here I was in bliss with the 32k context increase 3 days ago. 128k context? Absolutely insane. It feels like now the bottle neck in GPT workflows is no longer GPT, but instead its the wallet! Such an amazing time to be alive.

It’s insane because it makes no sense. When you read a book you don’t remember the last 100,000 words. It’s so wildly inefficient to do it that way.

Huh? By the time you finish reading a book you've forgotten the book?

Re: New models and developer products

#280

Post by dave1010uk »

gumballindie wrote: post #32
Zaheer wrote: post #15

The playbook OpenAI is following is similar to AWS. Start with the primitives (Text generation, Image generation, etc / EC2, S3, RDS, etc) and build value add services on top of it (Assistants API / all other AWS services). They're miles ahead of AWS and other competitors in this regard.

And just like amazon they will compete with their own customers. They are miles ahead in this regard as well since they basically take everyone’s digital property and resell it.

This is essentially the "Innovate - Leverage - Commoditise" strategy, which Simon Wardley (as in Wardley Mapping) explains:

https://blog.gardeviance.org/2014/03/understanding-ecosystem...

Post reply on HN