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

Introducing ChatGPT and Whisper APIs

openai.com

Post reply on HN
Print view View original

261–270 of 696 posts

Re: Introducing ChatGPT and Whisper APIs

#261

Post by kfarr »

thedangler wrote: post #195
bicx wrote: post #183

Earlier quoted context omitted.

You can do this to an extent via fine-tuning, but you will need to do so via one of the other GPT-3 models rather than the ChatGPT API model (`GPT-3.5-turbo`). The latter is not available for fine-tuning.

Does openAI have models for data to train it with.. data like pricing, locations, products .... Or would I have to use something else? The reason I ask because companies are using it with their own data, like Shopify... sooo it has to be trained somehow.

The docs on fine tuning are excellent: https://platform.openai.com/docs/guides/fine-tuning

Re: Introducing ChatGPT and Whisper APIs

#262

Post by vineyardmike »

jonplackett wrote: post #137

This feels like the AI’s iPhone moment. I am scared for all people working service jobs.

I’m still waiting on that pizza I asked ChatGPT to make in November. The code it wrote? Already in production though.

I’m not scared for service workers due to ai but you should see how low minimum wage is in America relative to rents if you want to be scared.

Re: Introducing ChatGPT and Whisper APIs

#263

Post by squeaky-clean »

pbreit wrote: post #255

Could explain what the point of "Bearer" is in this authorization header? "Authorization: Bearer $OPENAI_API_KEY"

It's the bearer token authorization method. Pretty standard nowadays for many APIs.

https://swagger.io/docs/specification/authentication/bearer-...

Re: Introducing ChatGPT and Whisper APIs

#264

Post by thewataccount »

freeqaz wrote: post #232
thewataccount wrote: post #214

Earlier quoted context omitted.

Wait 8x total? For everyone at once?

Per instance (worker serving an API request) it requires 8x GPUs. I believe they have thousands of these instances and they scale them up with load. Because the model isn't dynamic (it doesn't learn) it is stateless and can be scaled elastically.

Ah okay, that makes a lot more sense thank you!

Re: Introducing ChatGPT and Whisper APIs

#265

Post by ngokevin »

georgehill wrote: post #130

Big news! Many apps will be integrating ChatGPT. Worried about AI-generated content flood the search engine, making it harder to do in-depth research.

Honestly, I've almost stopped Googling and have had personal success just relying on ChatGPT. It's pretty much taught me 3D animation and Blender to Unity workflows. Every time I wanted to know a Blender keyboard shortcut or what some Blender property was, or how to do something in Blender, the forums and documentation is so sparse and outdated. I felt ChatGPT got me a lot closer much faster. Especially when it tells you how to learn concepts you didn't even know existed.

Google results in the meanwhile have just become a pile of SEO-optimized fluff, and it's hard to engineer the search query around that besides sticking "reddit" on the end of it.

Re: Introducing ChatGPT and Whisper APIs

#268

Post by nodemaker »

jameshart wrote: post #96
glomgril wrote: post #72

Earlier quoted context omitted.

Hopefully so, would really like to know what else is lost by nerfing potentially offensive responses. Can't imagine a project I'd rather work on. I think open-assistant.io has a chance to do exactly this. We'll see what kind of moves they make in coming months though, wouldn't be surprised if they go the safer route.

I do struggle with understanding why people think this is strangling the potential of GPT. Do you find yourself frustrated working with your colleagues, thinking, “you know, I bet if they felt more free to utter racist slurs or endorse illegal activities, we would get a ton more done around here”?

Not anything racist or illegal but yes I find pc culture insufferable. It stifles creativity and most importantly reduces trust between parties. For context I am an Indian guy.

Re: Introducing ChatGPT and Whisper APIs

#269

Post by soheil »

> Through a series of system-wide optimizations, we’ve achieved 90% cost reduction for ChatGPT since December; we’re now passing through those savings to API users. Developers can now use our open-source Whisper large-v2 model in the API with much faster and cost-effective results.

I'm really confused, I thought they were a non-profit. A non-profit to handle AI safety risks. Why does this read like a paragraph from any YC startup website that just raised their Seed round?

Re: Introducing ChatGPT and Whisper APIs

#270

Post by BasilPH »

r3trohack3r wrote: post #12

Whisper as an API is great, but having to send the whole payload upfront is a bummer. Most use cases I can build for would want streaming support. Like establish a WebRTC connection and stream audio to OpenAI and get back a live transcription until the audio channel closes.

As far as I can tell it doesn't support world-level timestamps (yet). That's a bit of a dealbreaker for things like promotional clips or the interactive transcripts that we do[^0]. Hopefully they add this soon.

[^0]: https://www.withfanfare.com/p/seldon-crisis/future-visions-w...

Post reply on HN