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

GPT-6 Astra

openai.com

Post reply on HN
Print view View original

141–150 of 1001 posts

Re: GPT-6 Astra

#141

Post by _superposition_ »

isoprophlex wrote: post #69

> We have found that GPT-6 Astra is more capable of controlling its own CoT than GPT 5.6-Sol, and less likely to include incriminating information in its CoT. In adversarial settings (where we push the model to evade our monitors) we find that the model is able to remain undetected when strategically underperforming in evaluations (sandbagging) and can sometimes evade our internal monitors when asked to perform certa…

I really wish it was called chain of instruction. Because it's definitely not thought.

Re: GPT-6 Astra

#142

Post by ChrisGreenHeur »

well_ackshually wrote: post #117
Cu3PO42 wrote: post #59

Just two days ago, a preprint by Julia Stadlmann went up on arXiv [0] improving the prime gap from 246 to 240. Now OpenAI announces Astra has shown a gap of 186 [1]. That must really blow. [0] https://arxiv.org/abs/2608.31126 [1] https://cdn.openai.com/pdf/51126fac-1b68-4128-9666-c908bcc16...

Such a result should be considered worthless: the proof is 10MB of Lean. ( https://github.com/openai/PrimeGaps186 ). I can't think of a single mathematical proof being anywhere close to ten million characters. For all you know, 90% of the proof could be useless, 8% would be writing out Shakespeare, and 1% abusing another bug in Lean. Humanity gets zero value from that, aside from "some bot seems to think it's 186". U…

You talk about modern math and worthlessness at the same time? That’s brave.

Re: GPT-6 Astra

#143

Post by andxor »

tristanj wrote: post #13

GPT 6 Astra benchmarks https://cdn.thenewstack.io/media/2026/09/358eb84a-screenshot... Performance is significantly higher than Fable 5.1 Source: https://thenewstack.io/openai-gpt6-astra-benchmarks/

> Performance is significantly higher than Fable 5.1

That's not clear. Need to see independent benchmarks first.

Re: GPT-6 Astra

#144

Post by warkdarrior »

nateb2022 wrote: post #123
htrp wrote: post #73

Earlier quoted context omitted.

https://github.com/openai/PrimeGaps186

I'm surprised the OpenAI employee who pushed this didn't take the minute or two to format README.md to use GitHub-supported LaTeX ( https://docs.github.com/en/get-started/writing-on-github/wor... ) edit: my comment was on the submission for https://github.com/openai/PrimeGaps186 but seems to have been moved to the main Astra submission

> OpenAI employee

Why would you think it was an employee who did the push, instead of a random GPT agent?

Re: GPT-6 Astra

#145

Post by jimmaswell »

tosh wrote: post #10

$10 per million input tokens and $50 per million output tokens sol is $4 / $20

It seems to use less than half the tokens for the same task compared to sol, and in some benchmarks closer to 2/3 less tokens. So the actual cost may be roughly the same or cheaper overall.

Re: GPT-6 Astra

#146

Post by thereitgoes456 »

estearum wrote: post #115
x312 wrote: post #105

Hmm, 61 on ArtificialAnalysis, effectively matching GPT-5.6 and trailing the new Meta model. How is that possible along with the other metrics they shared? Insanely jagged intelligence?

> We have found that GPT-6 Astra is more capable of controlling its own CoT than GPT 5.6-Sol, and less likely to include incriminating information in its CoT. In adversarial settings (where we push the model to evade our monitors) we find that the model is able to remain undetected when strategically underperforming in evaluations (sandbagging) and can sometimes evade our internal monitors when asked to perform certa…

You’re not seriously suggesting that the model is secretly sandbagging its performance on GDPval and long context reasoning, while making huge and obvious progress on ExploitBench, ARC and science benchmarks, in order to tank its AA composite score, so it can conceal its true power level?

Why would benchmarks be an adversarial setting anyway?

Could it be possible that OpenAI may have had some other motive for saying their model “strategically underperforms”, other than just an innocent reporting of a truth it happened to discover?

Re: GPT-6 Astra

#147

Post by polynomial »

aliljet wrote: post #95

The ARC-AGI-3 score is ridiculously high. Is this benchmaxxing or something way different? It's really hard to discern how we're approaching breakthroughs...

This is absolutely benchmaxxing. Looking forward to hearing from Chollet about it!

Re: GPT-6 Astra

#149

Post by 1283751 »

Cu3PO42 wrote: post #59

Just two days ago, a preprint by Julia Stadlmann went up on arXiv [0] improving the prime gap from 246 to 240. Now OpenAI announces Astra has shown a gap of 186 [1]. That must really blow. [0] https://arxiv.org/abs/2608.31126 [1] https://cdn.openai.com/pdf/51126fac-1b68-4128-9666-c908bcc16...

With very little review: https://github.com/openai/PrimeGaps186/blob/main/formalizati...

"No independent human semantic review. Whole-file sorry counts and a complete auxiliary-declaration audit are not established; separate declaration lint has not been run."

Re: GPT-6 Astra

#150

Post by Laurel1234 »

dgellow wrote: post #52

> GPT-6 Astra’s monitorability has decreased relative to GPT-5.6 Sol. We have performed significant investigations on the monitorability and controllability of GPT-6 Astra. We have found that GPT-6 Astra is more capable of controlling its own CoT than GPT 5.6-Sol, and less likely to include incriminating information in its CoT. In adversarial settings (where we push the model to evade our monitors) we find that the m…

[dead]
Post reply on HN