Liz Anthony explains why founder-led content consistently outperforms creator ads at Halfdays, sharing how lo-fi videos from the founders drove significantly higher ROAS and became a repeatable growth lever heading into Black Friday.

When the Community energy outperforms the market hype you end up with #GenuineUndead 🚀

R.I.P. basic prompting. MIT just dropped a technique that makes ChatGPT reason like a team of experts instead of one overconfident intern. It’s called “Recursive Meta-Cognition” and it outperforms standard prompts by 110%. Here’s the prompt (and why this changes everything) 👇

🧵 AI just changed everything for builders in Kenya. Here's what @MIT's Dave Dixon shared at #KenyaAIHackathon: AI scores 87.7% on PhD-level questions Outperforms doctors in diagnosis 300x cheaper in 12 months You have expertise in your pocket most don't know exists. 🧵👇🏿

True leadership does not eliminate opponents, it outperforms them. #ImranKhanIllegallyDetained @TeamiPians

👏 Kudos to Pracheta and team! 🎉 A clever galvanic displacement + operando activation strategy led to a Low-Pt HER electrode that outperforms bulk Pt with 133× less Pt. A major milestone in our No-Pt / Low-Pt self-supported electrode quest. https://t.co/0OrJItoKPa

Deadpool outperforms Phoenix trailer in less than 3 days😭

Kling 2.6 Motion Control is absolutely wild. It replaces characters with film-level precision — keeping motion, lighting, and expressions perfectly accurate. Easily outperforms Wan 2.2-Animate, Act-Two, and DreamActor 1.5 in real-world tests. Full tutorial 👇

Trump (and right wingers in Britain & EU) often claim that the USA economy outperforms “sclerotic” Europe. A more comprehensive comparison gives a different perspective about the quality of life in each:

Risk–reward is the real edge. A 30% win rate with 1:5 R:R outperforms a 70% win rate at 1:1. Professionals focus less on being right and more on making more when they are right. That asymmetry is what traders like Paul Tudor Jones have always emphasized.

Huge update for $WOM: We just integrated follower intelligence from @frontrunpro. We now have 11M+ smart-follower datapoints mapped into our scoring & reputation models. Combined with our ML layer trained on over 10M tweets, WOM now outperforms EVERY InfoFi platform we’ve

@sama The API team is doing a bad job, because you intend to remove GPT-4o-latest from API on Feb 17. GPT‑4o outperforms GPT‑5.2 in creative writing – by orders of magnitude. You said you had no plans to sunset 4o. You lied. #keep4o #keep4oAPI

GPT-5.2 beats human experts at their own jobs. OpenAI's new model ties or outperforms professionals on 70.9% of real work tasks — at 11x the speed and <1% the cost. The 30% it can't do is probably "send a Slack that sounds like you actually read the doc."

Just got around to reading the FineVision blog post from @huggingface today. Interesting findings: - FineVision outperforms baselines by 12-51% after sufficient training - FineVision contains least test contamination at 1%, versus 2-3% in baselines - Merging questions per image

Once again, $MANYU outperforms the macro market, ranking top 7 across ALL animal meme tokens on @CoinMarketCap and emerging as the highest gainer across the board. Their recent exposure on @GIPHY and growing virality marks the beginning of a Web3 → Web2 takeover, only preceded

LTX 2 uses gradient estimate euler denoising as described here: https://t.co/Gxmo66JYhj for faster/better sampling. The paper says it outperforms UniPC and DPM++ on FID.

The GLM-Image model (@Zai_org) was recently released and outperforms Nanobana 2.0 and GPT Image 1 on the CVTG-2k and LongText benchmarks. Compare the numbers: >CVTG-2k 1. Model: GLM-Image NED:0.9557 CLIPScore: 0.7877 2. Model: GPT Image 1 [High] NED:0.9478 CLIPScore: 0.7982

[AUDIO] Thursday Preview Call: Commodity FX outperforms.

Imagine being in the trenches since 2016 meanwhile AI: $OPUS Claude exists for 10 days and outperforms everything.