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

521–530 of 909 posts

Re: Open models by OpenAI

#521

Post by seany »

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…

Jail breaking then running censored questions. Like diy fireworks, or analysis of papers that touch "sensitive topics", nsfw image generation the list is basically endless.

Re: Open models by OpenAI

#522

Post by nfRfqX5n »

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…

You’re asking the biggest group of people who would want to do this

Re: Open models by OpenAI

#523

Post by jdiff »

pxc wrote: post #506
nojito wrote: post #494

Earlier quoted context omitted.

Why does it need knowledge when it can just call tools to get it?

Right... knowledge is one of the things (the one thing?) that LLMs are really horrible at, and that goes double for models small enough to run on normal-ish consumer hardware. Shouldn't we prefer to have LLMs just search and summarize more reliable sources?

Even large hosted models fail at that task regularly. It's a silly anecdotal example, but I asked the Gemini assistant on my Pixel whether [something] had seen a new release to match the release of [upstream thing].

It correctly chose to search, and pulled in the release page itself as well as a community page on reddit, and cited both to give me the incorrect answer that a release had been pushed 3 hours ago. Later on when I got around to it, I discovered that no release existed, no mention of a release existed on either cited source, and a new release wasn't made for several more days.

Re: Open models by OpenAI

#524

Post by mekpro »

coltonv wrote: post #393
hrpnk wrote: post #387

Earlier quoted context omitted.

With LM Studio you can configure context window freely. Max is 131072 for gpt-oss-20b.

Yes but if I set it above ~16K on my 32gb laptop it just OOMs. Am I doing something wrong?

try enable flash attention and offload all layer to GPU

Re: Open models by OpenAI

#525

Post by mekpro »

GodelNumbering wrote: post #372
simonw wrote: post #338

Just posted my initial impressions, took a couple of hours to write them up because there's a lot in this release! https://simonwillison.net/2025/Aug/5/gpt-oss/ TLDR: I think OpenAI may have taken the medal for best available open weight model back from the Chinese AI labs. Will be interesting to see if independent benchmarks resolve in that direction as well. The 20B model runs on my Mac laptop using less than 15GB…

> The 20B model runs on my Mac laptop using less than 15GB of RAM. I was about to try the same. What TPS are you getting and on which processor? Thanks!

i got 70 token/s on m4 max

Re: Open models by OpenAI

#526

Post by ninjin »

maxloh wrote: post #245

> We introduce gpt-oss-120b and gpt-oss-20b, two open-weight reasoning models available under the Apache 2.0 license and our gpt-oss usage policy. [0] Is it even valid to have additional restriction on top of Apache 2.0? [0]: https://openai.com/index/gpt-oss-model-card/

> Is it even valid to have additional restriction on top of Apache 2.0?

You can legally do whatever you want, the question is whether you will then for your own benefit be appropriating a term like open source (like Facebook) if you add restrictions not in line with how the term is traditionally used or if you are actually be honest about it and call it something like "weights available".

In the case of OpenAI here, I am not a lawyer, and I am also not sure if the gpt-oss usage policy runs afoul of open source as a term. They did not bother linking the policy from the announcement, which was odd, but here it is:

https://huggingface.co/openai/gpt-oss-120b/blob/main/USAGE_P...

Compared to the wall of text that Facebook throws at you, let me post it here as it is rather short: "We aim for our tools to be used safely, responsibly, and democratically, while maximizing your control over how you use them. By using OpenAI gpt-oss-120b, you agree to comply with all applicable law."

I suspect this sentence still is too much to add and may invalidate the Open Source Initiative (OSI) definition, but at this point I would want to ask a lawyer and preferably one from OSI. Regardless, credit to OpenAI for moving the status quo in the right direction as the only further step we really can take is to remove the usage policy entirely (as is the standard for open source software anyway).

Re: Open models by OpenAI

#527

Post by mplewis »

dongobread wrote: post #435
teitoklien wrote: post #408

Earlier quoted context omitted.

for now, i wouldnt rank any model from openai in coding benchmarks, despite all the false messaging they are giving, almost every single model openai has launched even the high end o3 expensive models are absolutely monumentally horrible at coding tasks. So this is expected. If its decent in other tasks, which i do find openai often being better than others at, then i think its a win, especially a win for the open so…

It is absolutely awful at writing and general knowledge. IMO coding is its greatest strength by far.

Sure sounds like they're not good at anything in particular, then.

Re: Open models by OpenAI

#529

Post by airtonix »

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…

[dead]

Re: Open models by OpenAI

#530

Post by mhitza »

mekpro wrote: post #525
GodelNumbering wrote: post #372

Earlier quoted context omitted.

> The 20B model runs on my Mac laptop using less than 15GB of RAM. I was about to try the same. What TPS are you getting and on which processor? Thanks!

i got 70 token/s on m4 max

That M4 Max is really something else, I get also 70 tokens/second on eval on a RTX 4000 SFF Ada server GPU.
Post reply on HN