149 comments

[ 2.3 ms ] story [ 89.0 ms ] thread
Open weight models from OpenAI with performance comparable to that of o3 and o4-mini in benchmarks… well, I certainly wasn’t expecting that.

What’s the catch?

Ha. Secure funding and proceed to immediately make a decision that would likely conflict viscerally with investors.
Text only, when local multimodal became table stakes last year.
Listed performance of ~5 points less than o3 on benchmarks is pretty impressive.

Wonder if they feel the bar will be raised soon (GPT-5) and feel more comfortable releasing something this strong.

Wow, today is a crazy AI release day:

- OAI open source

- Opus 4.1

- Genie 3

- ElevenLabs Music

So this confirms a best-in-class model release within the next few days?

From a strategic perspective, I can't think of any reason they'd release this unless they were about to announce something which totally eclipses it?

You hit the nail on the head!!!
Disclamer: probably dumb questions

so, the 20b model.

Can someone explain to me what I would need to do in terms of resources (GPU, I assume) if I want to run 20 concurrent processes, assuming I need 1k tokens/second throughput (on each, so 20 x 1k)

Also, is this model better/comparable for information extraction compared to gpt-4.1-nano, and would it be cheaper to host myself 20b?

Meta's goal with Llama was to target OpenAI with a "scorched earth" approach by releasing powerful open models to disrupt the competitive landscape. Looks like OpenAI is now using the same playbook.
Is there any details about hardware requirements for a sensible tokens per second for each size of these models?
I'm disappointed that the smallest model size is 21B parameters, which strongly restricts how it can be run on personal hardware. Most competitors have released a 3B/7B model for that purpose.

For self-hosting, it's smart that they targeted a 16GB VRAM config for it since that's the size of the most cost-effective server GPUs, but I suspect "native MXFP4 quantization" has quality caveats.

Please don't use the open-source term unless you ship the TBs of data downloaded from Anna's Archive that are required do build it yourself. And dont forget all the system prompts to censor the multiple topics that they don't want you to see.
Running a model comparable to o3 on a 24GB Mac Mini is absolutely wild. Seems like yesterday the idea of running frontier (at the time) models locally or on a mobile device was 5+ years out. At this rate, we'll be running such models in the next phone cycle.
GPQA Diamond: gpt-oss-120b: 80.1%, Qwen3-235B-A22B-Thinking-2507: 81.1%

Humanity’s Last Exam: gpt-oss-120b (tools): 19.0%, gpt-oss-120b (no tools): 14.9%, Qwen3-235B-A22B-Thinking-2507: 18.2%

So 120B was Horizon Alpha and 20B was Horizon Beta?
Shameless plug: if someone wants to try it in a nice ui, you could give Msty[1] a try. It's private and local.

[1]: https://msty.ai

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.
The repeated safety testing delays might not be purely about technical risks like misuse or jailbreaks. Releasing open weights means relinquishing the control OpenAI has had since GPT-3. No rate limits, no enforceable RLHF guardrails, no audit trail. Unlike API access, open models can't be monitored or revoked. So safety may partly reflect OpenAI's internal reckoning with that irreversible shift in power, not just model alignment per se. What do you guys think?
Exciting as this is to toy around with...

Perhaps I missed it somewhere, but I find it frustrating that, unlike most other open weight models and despite this being an open release, OpenAI has chosen to provide pretty minimal transparency regarding model architecture and training. It's become the norm for LLama, Deepseek, Qwenn, Mistral and others to provide a pretty detailed write up on the model which allows researchers to advance and compare notes.

Can't wait to see third party benchmarks. The ones in the blog post are quite sparse and it doesn't seem possible to fully compare to other open models yet. But the few numbers available seem to suggest that this release will make all other non-multimodal open models obsolete.
I dont see the unsloth files yet but they'll be here: https://huggingface.co/unsloth/gpt-oss-20b-GGUF

Super excited to test these out.

The benchmarks from 20B are blowing away major >500b models. Insane.

On my hardware.

43 tokens/sec.

I got an error with flash attention turning on. Cant run it with flash attention?

31,000 context is max it will allow or model wont load.

no kv or v quantization.

What a day! Models aside, the Harmony Response Format[1] also seems pretty interesting and I wonder how much of an impact it might have in performance of these models.

[1] https://github.com/openai/harmony