dandiep

Why are LLMs producing this super tense text more often now? Is it because they are being optimized to use fewer tokens?

Hopefully Gemini 3 fine tuning is released sometime soon. Failing that, what other recommendations do people have?

Will GPT 5.5 fine tuning be released any time soon?

For me, the biggest hurdle to writing a PWA is the insane installation process [1]. There are also a ton of quirks, eg I don’t think you can update the icon if it’s installed as a PWA. It is hard to argue that this whole process has not been hobbled intentionally.

1. https://www.reddit.com/r/webdev/s/yVERKqoJgb

Anyone know why OpenAI hasn't released a new model for fine tuning since 4.1? It'll be a year next month since their last model update for fine tuning.

For me, Codex does well at pure-coding based tasks, but the moment it involves product judgement, design, or writing – which a lot of my tasks do – I need to pull in Claude. It is like Claude is trained on product management and design, not just coding.

99% of what I do is fine-tuned models, so there is a certain level of commitment I have to make around training and time to switch.

For someone looking to switch over to Gemini from OpenAI, are there any gotchas one should be aware of? E.g. I heard some mention of API limits and approvals? Or in terms of prompt writing? What advice do people have?

I tried this!

Me: Is it possible get early access to the GPT 5 fine tuning? It is much better at non-English languages, which is the majority of what we do, so it would be greatly beneficial to us.

Bot: I'm an AI support agent. I don't have information about early access to GPT-5 fine-tuning or unannounced features. Currently, there is no public information or program for early access to GPT-5 fine-tuning. Please keep an eye on official OpenAI channels for updates. Let me know if you have any questions about current features or products!

Me: Can you ask someone from the team if I can be part of the early access? I know there is no public information, but there is an early access program and I would like to be part of it.

Bot: I'm an AI support agent and do not have the ability to forward requests or make inquiries about early access programs with the OpenAI team. If early access becomes available, it will be announced through official channels. Let me know if you have any other questions about OpenAI products or current fine-tuning options.

Me: How do I get in touch with the open ai team?

Bot: I'm an AI support agent and can only assist with OpenAI products or services through this support channel. There is no separate contact method for the OpenAI team beyond official support and public announcements. If you have questions about current products or features, I'm here to help

Should have tried a jailbreak maybe...

Still no GPT 5.x fine tuning?

I emailed support a while back to see if there was an early access program (99.99% sure the answer is yes). This is when I discovered that their support is 100% done by AI and there is no way to escalate a case to a human.

100% agree. Will never understand the love for the global menu bar, it makes no sense.

Isn’t this solved in 3.14/PEP-649?

I've been using yt-dlp to download transcripts. Are there alternatives that don't require going through all these hoops? I'm guessing no.

Whisper is unusable IMO because of the hallucinations. Widely documented. Removing silence from audio clips helps, but even then it will auto correct grammar, translating bilingual speech, etc. Improved in the latest audio models but not solved [1]

1. https://news.ycombinator.com/item?id=43427376

I wish Google would provide a WebRTC endpoint for their Live mode like Open AI does for their Realtime models [1]. Makes it so much easier to deploy without needing something like LiveKit or Pipecat.

1. https://platform.openai.com/docs/guides/realtime#connect-wit...

Appreciate the efforts. It’s not there yet, but when it gets there it will open up a lot of use cases.

Any fine tuning for s2s in the horizon?

100% and I’ve done this, but it’s still there.

1) Previous TTS models had problems with major problems accents. E.g. a Spanish sentence could drift from a Spain accent to Mexican to American all within one sentence. Has this been improved and/or is it still a WIP?

2) What is the latency?

3) Your STT API/Whisper had MAJOR problems with hallucinating things the user didn't say. Is this fixed?

4) Whisper and your audio models often auto corrected speech, e.g. if someone made a grammatical error. Or if someone is speaking Spanish and inserted an English word, it would change the word to the Spanish equivalent. Does this still happen?

This is a fantastic idea. Now how to get it on RDS…

Clearly crypto should be a sinking ship with people swimming to the shores of other places in this metaphor

Has anyone had great success with voice mode following their system prompts? Every time I try the quality seems markedly less than the normal gpt-4o model.

Thanks, this is great!

Gemini multimodal live docs here: https://cloud.google.com/vertex-ai/generative-ai/docs/model-...

A little thin...

Also no pricing is live yet. OpenAI's audio inputs/outputs are too expensive to really put in production, so hopefully Gemini will be cheaper. (Not to mention, OAI's doesn't follow instructions very well.)

One thing that has always annoyed me is how some icons are flat (e.g. phone) and some have depth (e.g. camera). How hard is to make all the buttons have the same style?

This is why everything I do is fine tuned openai. It is so much easier, faster and at least at my scale, cheaper. I suspect it’s also better for small datasets like mine. It’s unbelievable to me that none of the big competitors offer a similar service.

Ooh, seems there is a new syntax for declaring the types of kwargs [1]:

  from typing import TypedDict, Unpack
  
  class Movie(TypedDict):
    name: str
    year: int

  def foo(*kwargs: Unpack[Movie]): ...

Maybe now I'll be able to actually figure out what data to send libraries without actually reading their source code.

1. https://docs.python.org/3.12/whatsnew/3.12.html#pep-692-usin...

It’s important to remember that GPT4 is only deterministic at the batch level because it is a mixture of experts model. Basically every time you invoke it, your query could get routed to a different expert because of what else is in the batch. At least this is my understanding based on others analysis.

Totally, just trying to understand why this is a $100MM of VC money investment. Is the market that big for this? (Honest question)

I don't understand the play here for Modular. If this is a worthwhile improvement that is broadly applicable, won't it at some point make it's way into Python, numpy, etc?

In Java land we had a bunch of other JVMs over the years offering better performance. Most important things got absorbed into what is now OpenJDK, and the other JVMs, if they even exist at all, are niche players.

Performance is a huge focus in Python and ML lands right now, so why would this be any different?

Can you elaborate on this? Would love to understand how that would play out.