Only Tue, September 22, 2026 · All dates →
Here’s my outlook on humanoid robots. There are four stages of maturity, and each is gated by progress in the stage before it First, you have to build great humanoid hardware Second, that hardware needs to work with a whole-body, AI-first architecture These f…
The goal of building any technology, AI included, should be bettering human lives and society. [Quoting @BloombergTV]: “Any threat to human society, including existential, is within ourselves,” says World Labs Technologies CEO Fei-Fei Li as she discusses the…
wtf is this GM hikaru nakamura, no 2 ranked chess player in the world, loving muse?? senpai!! 🙇♂️ if it's good enough for him it's probably good enough for the rest of us tbh [Quoting @GMHikaru]: Muse is the real deal. I have yet to see any of the other LLM…
I am sorry, Opus 5.5 is nowhere near Fable 5.1 Planning to stick with Fable 5.1 for all hard-coding problems. It continues to be brilliant 🤷♀️
We're sharing new research on our post‑training approach, which teaches the Perplexity Computer agent to learn from real user sessions by imitating good trajectories and explicitly correcting avoidable mistakes like bad tool calls (even when the overall traje…
People are loving 5.5. Congrats to Anthropic!
GPT-6 Luna is half the price of 5.6 Luna, which was already an astonishingly cheap model given how capable it is Luna is my favorite model for building product features thanks to its cost (and speed) [Quoting @OpenAI]: Please welcome GPT-6 Sol and GPT-6 Luna…
pretty psyched about this one— expedia 🤝 muse we've seen travel be a breakout use case for muse, and excited to partner with expedia to make it seamless to go anywhere ✈️ [Quoting @Expedia]: Here at Expedia, we love planning trips. Soon, your personal AI age…
Enterprises should own their intelligence. That means bringing AI to their sensitive data while protecting both business data and proprietary models. NVIDIA’s @j_boitano and @VAST_Data Renen Hallak discuss how NVIDIA Confidential Computing makes this possible.
LOL! As predicted, the frontier model tsunami just started GPT 5.6 Sol and Luna at 50% off standard prices! Luna just became competitive with DeepSeek. Expect a massive spike in adoption 😲
Well said. [Quoting @theo]: Hot take: this is what pacing looks like. None of today's releases were Astra or Fable tier. This is intentional. The point of "pacing" isn't to stop iteration and improvement. The goal is to prevent the development bigger models f…
GPT-6 Sol is now available to all Perplexity users. On our Wide-And-Deep-Research (WANDR) evals, it outperforms Opus 5 at one-fifth the price. Sol will become the “Light” Effort orchestrator for Computer users, while Astra remains the orchestrator for "High"…
The new GPT-6-Sol and GPT-6-Luna are now available in Hermes Agent in Nous Portal and OpenRouter, soon through Codex subscriptions as well! [Quoting @OpenAI]: Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe. GPT-6 Sol and Luna build on the advan…
GPT-6 Sol and Luna — faster, more affordable models with the advances behind Astra’s SOTA performance in professional work, factuality, coding, computer use, and alignment: [Quoting @OpenAI]: Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe. GPT-…
Startups are naturally good at this; it is hard to keep a bigger company good at this and i think an underexplored space. [Quoting @prd_008]: OpenAI's superpower is the ability to swiftly assemble an empowered group of highly capable people to work on the mos…
We want the OpenAI API to feature the best model at every price point and to be the best at every modality (text, code, image, video, etc). And then we want you all to come up with great ideas and build them and to get to be happy users. The best ideas will c…
Especially compared by per-task pricing, which is the metric that should matter, I don't think there is anything competitive anywhere in the market. We want people to be able to use tons of AI; it is important to being able to explore this new renaissance in…
GPT-6 Sol and Luna are out, and they are better AND 50% cheaper than 5.6. Luna is now $0.10 input / $0.50 output per 1M tokens. This is on top of the 80% price cut to Luna we made at the end of July. Output went from $6 -> $0.50 within two months. [Quoting…
GPT-6 Sol and Luna are big improvements on intelligence, alignment, work output, coding, computer use, and more over their 5.6-family predecessors. They are also half the price per token, and even less per task!
GPT-6 Sol and Luna are great models but also these characters are so cute [Quoting @OpenAI]: Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe. GPT-6 Sol and Luna build on the advances behind GPT-6 Astra, bringing much of its strengths into faster…
Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe. GPT-6 Sol and Luna build on the advances behind GPT-6 Astra, bringing much of its strengths into faster and more affordable models to support work at scale. We’ve also made caching and inference m…
Opus 5.5 feels like chatting with a way smarter Opus 4.6. I’m so happy.
its interesting how claude manually wielding the brush leads to a very different form of creativity than diffusion based models (and it shows how deeply this generation of claude can _see_) [Quoting @jkeatn]: for the past few months i've been asking our model…
/btw to initiate side chats inside a Computer session with a shared snapshot of the context. [Quoting @AskPerplexity]: We launched side chat in Computer. Ask a quick question about a Computer task in a separate, read-only side chat while the main task keeps g…
Frontier: pacing Pulling up the ladder: accelerating [Quoting @M1Astra]: Claude Opus 5.5 will be the first Opus meant to fall back to a less capable model for "a small set of capabilities related to the development of frontier LLMs, such as kernel development…
Claude Opus 5.5 is now available in Cursor! It's the new top model on CursorBench at 57.8% (Max) and costs 40% less per task than Opus 5.
As part of our efforts to pace the frontier, we’re committed to supporting independent assessments with deep levels of access across training, evaluation, and deployment. That access should enable third party assessors to challenge our assumptions, identify r…
Whenever I see a model top this chart, it's normally a bit of a red flag. Would love to be proven wrong! [Quoting @synthwavedd]: As expected, unfortunately the only bad thing about the new Opus is it's a token-gobbling monster. Uses the most tokens per Intell…
The tweet below got enormous pushback at the time I posted it, more than 5 years ago. But this prediction is looking obvious by the day now. [Quoting @fchollet]: Within 10-20 years, nearly every branch of science will be, for all intents and purposes, a branc…
Lots more coming soon. Today is just the start! [Quoting @mikeyk]: We’re also launching Claude Sonnet 5.5 and Haiku 5.5 in the coming weeks. These models will come with many of the same improvements to performance, efficiency, and safety. With that, happy bui…
The A.I. Daily