Though the smartness comes with a cost: all of the prompts that were written for the old realtime voice model now need to be revised for a more capable, and better instruction following, model.
LLMS
-
Analyzing the significance of OpenAI’s new Daybreak release
By
–
OpenAI 凌晨上了一个叫 Daybreak 的东西。4 小时 1M 浏览，不过讨论区基本没在讨论这到底是什么。
— 艾略特 (@elliotchen100) 12 mai 2026
整体看下来，关键不在 「OpenAI 发了一个新产品」，而在 「OpenAI 在 cyber AI 这个最敏感的话题上，终于跟 Anthropic 划了一条清晰的线」。
1. Daybreak 不是一个产品，是一把伞
Greg Brockman 自己… https://t.co/0qBIRMm8lCOpenAI dropped something called Daybreak in the wee hours of the morning. 4 hours in, 1M views, but the discussion section is basically not talking about what this thing actually is. Looking at it overall, the key isn't "OpenAI released a new product," but "OpenAI has finally
-
Frontier model writing: good style but weak spots and clichés
By
–
I think frontier model writing is good! It often has a sense of style & tone, variations in sentence structure & length, some great phrasing, etc But it also has some weak spots (fiction!) & clear tics. Mostly there is just far too much of it online which makes it all so cliche
-
Analyzing Incremental Progress and Moats in General Purpose LLMs
By
–
When is the last time a general purpose LLM (putting aside hybrid systems like Claude Code with special purpose symbolic harnesses) last completely blew away all competing prior models? GPT-4 relative to GPT 3.5? That’s what incremental change with no real moat looks like.
-
GPT-Realtime-2 voice model is smarter but lacks benchmarks
By
–
gpt-realtime-2 is a great voice model (with a typically bad OpenAI name). Voice models are natively processing speech, not transcribing it, so the intelligence of the model matters. The old voice model was GPT-4o level, this is much smarter (how smart? OpenAI gave no benchmarks)
-
Automating Travel Bookings with AI Agents
By
–
I put my flight preferences in my Cowork instructions, then let Opus get to work. It opened my browser, navigated a bunch of websites, and booked everything for me. The result: Cowork booked 8 flights and 5 hotels for me, while I was hacking on something else in Claude Code. It
-
Using Claude Cowork and Opus 4.7 for automated flight booking
By
–
I needed to book flights for a bunch of upcoming travel. As always, I used Claude Cowork to do it. In the past, Cowork has been decent at booking flights, but with Opus 4.7, for the first time ever, it 1-shotted it!
-
Ilya Sutskever on the extreme compute costs of AI development
By
–
And right before that, when @ilyasut explained the cost of all the compute: “You look at this number of dollars and you say ‘gosh that’s a lot of dollars.’”