X日报 · AI科技

♥️ GPT-5.6 is a major step forward for health, both at the frontier and at cost.

These models push the frontier of performance per dollar, bringing the best health intelligence to all. The smallest variant, GPT-5.6 Luna, evaluated at the lowest reasoning effort, outperforms GPT-5.5 at the highest reasoning effort–despite costing 25x less. The largest variant, GPT-5.6 Sol, sets a new high bar at cost.

Another especially cool result: physicians found fewer flaws in GPT-5.6 responses than physician-written responses.

We collected diverse tasks that remain difficult for recent OpenAI models, across patient-facing and clinician-facing use cases. We asked speciality-matched physicians to write responses to these tasks with unlimited time and web access. We then asked other physicians to compare responses side-by-side, blinded to their source. Physicians were asked to comment on areas of improvement across five axes: accuracy, communication, completeness, instruction following, and health decision helpfulness. We then reported the fraction of responses across sources rated perfectly across all axes, across 20,000 total axis ratings. GPT-5.6 Sol appeared strongest, although all GPT-5.6 models performed significantly better than physicians.

@gdb our benchmark shows that Sol ranks #1 is 6x more cost efficient than Fable across React/frontend work

BREAKING - OFFICIAL RESULTS: GPT-5.6 Sol by @OpenAI is 1st overall on Design Arena with an Elo of 1353.

This puts GPT-5.6 Sol above Claude Fable 5 by @AnthropicAI and in the same performance band as GLM 5.2 by @Zai_org on frontend design.

This is an 18-position and 60-point Elo leap from GPT-5.5.

GPT-5.6 Sol also establishes a new Pareto frontier for preference vs. speed, faster than any model at this performance.

Congratulations to the @OpenAI team on the launch!

GPT 5.6 Sol is Insane for web designers!

Prompt ↓ https://t.co/r8kWamEva4

our team's first week of having both sol and fable
the crazy thing is 30% of our cost was on fable https://t.co/8yVddGEuBg

GPT-5.6 Sol Ultra found me a solution to another Erdos problem not long after this one.

Problem #793 asks about asymptotic of a 2-primitive set. GPT again came up with an extremely short and elegant construction that enhances original Erdos method used to prove a weaker result.

AI has helped resolve an important question in statistics. In the area of multiple hypothesis testing, the goal of controlling the false discovery rate (FDR) has been introduced in a seminal paper by Benjamini and Hochberg (1995). They also introduced a method (the Benjamini-Hochberg or BH method) and proved it controls the FDR. This method has been widely adopted in modern high-throughput science, including in genomics, astronomy, economics, etc. The paper has has garnered more than 130,000 citations to date.

However Benjamini and Hochberg showed FDR control only when the data for the individual tests are *independent*. In practice, these data are often dependent; a good example is data on genetic variants due to linkage disequilibrium. Later work has focused on extending the validity of the BH procedure, e.g., to a form of positive dependence by Benjamini and Yekutieli (2001).

The question of when the BH procedure controls the FDR has remained open. Over the last twenty years, many authors, including Reiner-Benaim (2007), Kim and van de Wiel (2008), Benjamini (2010), Sarkar (2023), Sarkar and Zhang (2025), have conjectured that the BH procedure controls the FDR for two-sided tests using any correlated Gaussian data. These authors have presented both theoretical and empirical evidence supporting, but not directly showing, the conjecture.

With the help of AI (specifically GPT-5.6 Sol Pro), I have settled the question in the negative: The Benjamini-Hochberg procedure does *not* generally control the false discovery rate at the desired level for correlated two-sided Gaussian tests. This was done by exhibiting a Gaussian factor model for which, at a nominal level alpha=0.01, the false discovery rate is proved to be FDR>0.0104.

There is a lot of interesting commentary to be made:

1. This result should be of interest to everybody in the field of statistics. Emmanuel Candes of Stanford University once called the false discovery rate and the Benjamini-Hochberg procedure "one of the two most important developments in statistics after 1950" (the other being James-Stein shrinkage). The present conjecture is probably the most central question about FDR/BH that was unresolved to date.

2. GPT-5.6 one-shot the problem after 90 minutes of reasoning, whereas with 5.5 I was not able to solve it even after iterating with multiple parallel agents for perhaps 20 hours. So the capability improvement is quite real. Exciting times to live in!

3. The argument is not especially surprising, but it does combine an asymptotic approach (standard for FDR analysis, see e.g., Genovese and Wasserman, Efron, etc) with a numerical certificate in a way that would be pretty non-standard in the field. Once we have the specific example, then straightforward simulations also support that the false discovery rate is indeed higher than the nominal value (see attached fig).

4. The current degree of violation over the nominal level is relatively small (0.104 vs 0.1). So the importance of this result is mainly conceptual. The practical implications remain to be determined.

Overall, an exciting development! Preprint is available here (https://t.co/YgiwgDF2qr) and will be on arxiv tonight; supporting code is here (https://t.co/KZhj15qDXC).

GPT-5.6 Sol, Terra, and Luna from @OpenAI are now generally available on Amazon Bedrock. Three tiers of intelligence, from flagship reasoning to fast inference, running on Bedrock's next-generation inference engine built for high-performance, security, scale, and reliability. https://t.co/Hn2z5M2Jp2

CODEX SKILL THAT FINDS YOUR STARTUP’S FIRST CUSTOMERS!

I made a Codex skill that analyzes your startup and finds potential customers from real public signals.

Paste your startup URL while Codex defines your ideal customer, searches public discussions, qualifies each prospect, and generates a polished report with personalized outreach openers.

-> ideal customer profile analysis
-> public pain + buying signal research
-> evidence-backed prospect shortlist
-> fit, timing + reachability scores
-> original source links for every prospect
-> personalized outreach openers
-> polished HTML report
-> one-command install

Install: npx --yes codex-first-customer-finder-skill

100% open source.
Repo in Bio.

a lot of people haven’t realized how much more capable Chat has become in the last few days

how much of Codex is now available directly inside @ChatGPTapp on mobile and the web.

the Codex desktop app is great. I obviously live there.

but now, inside ChatGPT, you can tap Work on mobile or the web to hand off a task that runs in the cloud, no computer needed.
you can also tap Remote to pick up and continue work already running on your computer.

start at your desk, keep things moving on your phone, or kick off something new wherever you are.

agents shouldn’t care which screen you started from.

ChatGPT Sites is now in public beta.

You can turn a prompt, file, or rough idea into a dashboard, project tracker, report, prototype, or lightweight app:

build and edit directly in ChatGPT Work or Codex
test in a private preview publish and share with a URL

Rolling out across paid plans. Enterprise admins can control public publishing.

Try out the Visualize plugin (in preview) in Codex.

Some ideas click faster when you can see and interact with it.

Here’s a fun example where I asked Codex to build a planet simulator with different controls. 🌎 https://t.co/Ezzi9YJJve

Use @ChatGPTapp Work for your most important tasks ⚽️ https://t.co/hR2NHfC4aC

Thank you to the 7M active users who are now using Codex and ChatGPT Work.

We have added a banked reset to everyone's account to celebrate the milestone. You can apply the reset in the desktop app or on web and it will replenish the weekly usage for you.

Have fun out there.

Today, we are introducing Inkling.

Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available.

https://t.co/Ghebq5mG30

Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵

first open weight thinking machine model!! 975B total, 41B active trained on 45T tokens, 1M context, multimodal in

sliding window with a 5:1 ratio and 512 size, deepseek aux-free load balancing and 2 shared experts (usually people only use 1), actually curious why the model is less sparse than kimi (~4.2% vs 3.2%). they use a short convolution after k and v, output and ffn (see plot), muon (they cite manifold muon but mention weight decay so not sure), muP, and have a very nice RL scaling curve and chain of thought!

one very cool part of the release imo is how well their small variant (276B total, 12B active) performs compared to the big one. they mention they changed the pre-training data mix and recipe, very curious about those changes and to see them scaled up to ~1T (or more?) soon 👀

> "It is not the most performant model available today, closed or open. We trained Inkling for solid capabilities across the board rather than state-of-the-art performance in a single area, to serve as a foundation for the models we will train in the future."

also this is really refreshing to see in a model release, huge congrats :)

Inkling by @thinkymachines is now available on Modal, backed by a custom DFlash speculator for 67% higher throughput and interactivity.

Running on Modal Auto Endpoints with SGLang today. https://t.co/OxN7aJ9ieW

We're building AI that people and organizations can shape and make their own. AI should extend our will and judgment instead of neglecting it; enabling that is the technical challenge we are working to solve.
https://t.co/Bi558y4vqD