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

Open models by OpenAI

openai.com

Post reply on HN
Print view View original

631–640 of 909 posts

Re: Open models by OpenAI

#631

Post by Oreb »

staplers wrote: post #588
xrmagnum wrote: post #509

Earlier quoted context omitted.

It's striking how much of the AI conversation focuses on new use cases, while overlooking one of the most serious non-financial costs: privacy. I try to be mindful of what I share with ChatGPT, but even then, asking it to describe my family produced a response that was unsettling in its accuracy and depth. Worse, after attempting to delete all chats and disable memory, I noticed that some information still seemed to…

Worse, after attempting to delete all chats and disable memory, I noticed that some information still seemed to persist. Chatgpt was court ordered to save history logs. https://www.malwarebytes.com/blog/news/2025/06/openai-forced...

That only means that OpenAI have to keep logs of all conversations, not that ChatGPT will retain memories of all conversations.

Re: Open models by OpenAI

#632

Post by anticensor »

lukasgross wrote: post #86
dsco wrote: post #33

Does anyone get the demos at https://www.gpt-oss.com to work, or are the servers down immediately after launch? I'm only getting the spinner after prompting.

(I helped build the microsite) Our backend is falling over from the load, spinning up more resources!

Why isn't GPT-OSS also offered on the free tier of ChatGPT?

Re: Open models by OpenAI

#633

Post by anticensor »

BoorishBears wrote: post #363
redox99 wrote: post #215

Earlier quoted context omitted.

There's nothing new here in terms of architecture. Whatever secret sauce is in the training.

Part of the secret sauce since O1 has been accesss the real reasoning traces, not the summaries. If you even glance at the model card you'll see this was trained on the same CoT RL pipeline as O3, and it shows in using the model: this is the most coherent and structured CoT of any open model so far. Having full access to a model trained on that pipeline is valuable to anyone doing post-training, even if it's just to…

Its CoT is sadly closer to that sanitised o3 summaries than to R1 style traces.

Re: Open models by OpenAI

#634

Post by jona777than »

captainregex wrote: post #397
cco wrote: post #357

The lede is being missed imo. gpt-oss:20b is a top ten model (on MMLU (right behind Gemini-2.5-Pro) and I just ran it locally on my Macbook Air M3 from last year. I've been experimenting with a lot of local models, both on my laptop and on my phone (Pixel 9 Pro), and I figured we'd be here in a year or two. But no, we're here today. A basically frontier model, running for the cost of electricity (free with a rounding…

I’m still trying to understand what is the biggest group of people that uses local AI (or will)? Students who don’t want to pay but somehow have the hardware? Devs who are price conscious and want free agentic coding? Local, in my experience, can’t even pull data from an image without hallucinating (Qwen 2.5 VI in that example). Hopefully local/small models keep getting better and devices get better at running bigger…

One of my favorite use cases includes simple tasks like generating effective mock/masked data from real data. Then passing the mock data worry-free to the big three (or wherever.)

There’s also a huge opportunity space for serving clients with very sensitive data. Health, legal, and government come to mind immediately. These local models are only going to get more capable of handling their use cases. They already are, really.

Re: Open models by OpenAI

#635

Post by nixgeek »

rasmus1610 wrote: post #627
atlasunshrugged wrote: post #573

Earlier quoted context omitted.

Add government here too (along with all the firms that service government customers)

Add healthcare. Cannot send our patients data to a cloud provider

A ton of EMR systems are cloud-hosted these days. There’s already patient data for probably a billion humans in the various hyperscalers.

Totally understand that approaches vary but beyond EMR there’s work to augment radiologists with computer vision to better diagnose, all sorts of cloudy things.

It’s here. It’s growing. Perhaps in your jurisdiction it’s prohibited? If so I wonder for how long.

Re: Open models by OpenAI

#636

Post by 1123581321 »

robwwilliams wrote: post #507
int_19h wrote: post #496

Earlier quoted context omitted.

I tried 20b locally and it couldn't reason a way out of a basic river crossing puzzle with labels changed. That is not anywhere near SOTA. In fact it's worse than many local models that can do it, including e.g. QwQ-32b.

Well river crossings are one type of problem. My real world problem is proofing and minor editing of text. A version installed on my portable would be great.

Dozens of locally runnable models can already do that.

Re: Open models by OpenAI

#637

Post by 9rx »

int_19h wrote: post #496
cco wrote: post #357

The lede is being missed imo. gpt-oss:20b is a top ten model (on MMLU (right behind Gemini-2.5-Pro) and I just ran it locally on my Macbook Air M3 from last year. I've been experimenting with a lot of local models, both on my laptop and on my phone (Pixel 9 Pro), and I figured we'd be here in a year or two. But no, we're here today. A basically frontier model, running for the cost of electricity (free with a rounding…

I tried 20b locally and it couldn't reason a way out of a basic river crossing puzzle with labels changed. That is not anywhere near SOTA. In fact it's worse than many local models that can do it, including e.g. QwQ-32b.

I tried the two US presidents having the same parents one, and while it understood the intent, it got caught up in being adamant that Joe Biden won the election in 2024 and anything I do to try and tell it otherwise is dismissed as being false and expresses quite definitely that I need to do proper research with legitimate sources.

Re: Open models by OpenAI

#638

Post by lend000 »

cco wrote: post #357

The lede is being missed imo. gpt-oss:20b is a top ten model (on MMLU (right behind Gemini-2.5-Pro) and I just ran it locally on my Macbook Air M3 from last year. I've been experimenting with a lot of local models, both on my laptop and on my phone (Pixel 9 Pro), and I figured we'd be here in a year or two. But no, we're here today. A basically frontier model, running for the cost of electricity (free with a rounding…

For me the game changer here is the speed. On my local Mac I'm finally getting token counts that are faster than I can process the output (~96 tok/s), and the quality has been solid. I had previously tried some of the distilled qwen and deepseek models and they were just way too slow for me to seriously use them.

Re: Open models by OpenAI

#639

Post by ludwik »

xrmagnum wrote: post #509
captainregex wrote: post #397

Earlier quoted context omitted.

I’m still trying to understand what is the biggest group of people that uses local AI (or will)? Students who don’t want to pay but somehow have the hardware? Devs who are price conscious and want free agentic coding? Local, in my experience, can’t even pull data from an image without hallucinating (Qwen 2.5 VI in that example). Hopefully local/small models keep getting better and devices get better at running bigger…

It's striking how much of the AI conversation focuses on new use cases, while overlooking one of the most serious non-financial costs: privacy. I try to be mindful of what I share with ChatGPT, but even then, asking it to describe my family produced a response that was unsettling in its accuracy and depth. Worse, after attempting to delete all chats and disable memory, I noticed that some information still seemed to…

> Worse, after attempting to delete all chats and disable memory, I noticed that some information still seemed to persist.

I'm fairly sure "seemed" is the key word here. LLMs are excellent at making things up - they rarely say "I don't know" and instead generate the most probable guess. People also famously overestimate their own uniqueness. Most likely, you accidentally recreated a kind of Barnum effect for yourself.

Re: Open models by OpenAI

#640

Post by NitpickLawyer »

jlokier wrote: post #606
NitpickLawyer wrote: post #340

Earlier quoted context omitted.

Please see the detailed response to a sibling post. tl;dr; weights are not binaries.

"Compiled binaries" are just meant to be an example. For the purpose of whether something is open source, it doesn't matter whether something is a "binary" or something completely different. What matters (for all common definitions of open source): Are the files in "source form" (which has a definition), or are they "derived works" of the source form? Going back to Apache 2.0. Although that doesn't define "open sourc…

> "Source" form shall mean the PREFERRED FORM FOR MAKING MODIFICATIONS, including BUT NOT LIMITED TO software source code, documentation source, and configuration files.

Yes, weights are the PREFFERED FORM FOR MAKING MODIFICATIONS!!! You, the labs, and anyone sane modifies the weights via post-training. This is the point. The labs don't re-train every time they want to change the model. They finetune. You can do that as well, with the same tools/concepts, AND YOU ARE ALLOWED TO DO THAT by the license. And redistribute. And all the other stuff.

Post reply on HN