In partnership with
News of the day
1. Meta launches its new Muse Wonder AI models, demonstrating impressive performance across various benchmarks and offering competitive pricing. → Read more
2. Z.ai launches GLM-Legal, a specialized AI model for legal tasks, achieving a 19.17% pass rate on Harvey LAB, significantly outperforming competitors. → Read more
3. Google DeepMind is establishing a new robotics lab in the Boston area, expanding its hardware team to integrate advanced AI into physical systems. → Read more
4. NVIDIA RTX Spark laptops launch Oct. 16, featuring powerful local AI, enhanced creative apps, and new developer tools for a reinvented PC experience. → Read more
Our take
Hi Dotikers!
Meta just dropped a surprise on the AI world, and this one actually deserves your attention. The company unveiled its new Muse Wonder family, two models called Muse Wonder and Muse Wonder Full, built for frontier coding, long-running agents and advanced computer use. On paper, the benchmarks are impressive, with Muse Wonder Full outperforming several big-name rivals across multiple categories, and the pricing strategy is clearly designed to hurt the competition: Muse Wonder is available through a simple API, while the Full version sits behind a Muse Code subscription.
What makes this release interesting is the context. Meta has spent the last two years being treated as the open-source underdog of the AI race, generous with its models but rarely seen as the performance leader. With Muse Wonder, the message changes completely. The company is no longer content giving away tools for the community; it wants to compete head-on in the premium agent and coding market, where OpenAI and Anthropic have been comfortably collecting developer subscriptions. Turns out Zuckerberg did not spend billions on GPUs just to generate better birthday posts.
The cost-per-task data and intelligence index rankings Meta published should obviously be taken with a grain of salt, since every lab conveniently wins its own benchmarks. But the direction is unmistakable: the battle is shifting from raw chatbot intelligence to agents that can work for hours, write production code and operate a computer like a junior employee. That is where the real money is, and Meta knows it. If the independent evaluations confirm even half of these claims, the pricing pressure alone will force everyone else to react, and that is excellent news for all of us who actually pay these bills.
Grab a coffee, this edition dives into what Muse Wonder really changes.
G.
Blu Dot surpasses 2,000% ROAS with self-serve CTV ads
Home furniture brand Blu Dot blew up on CTV with help from Roku Ads Manager. Here’s how:
After a test campaign reached 211,000 households and achieved 1,010% ROAS, the brand went all in to promote its annual sales event. It removed age and income constraints to expand reach and shifted budget to custom audiences and retargeting, where intent was strongest.
The results speak for themselves. As Blu Dot increased their investment by 10x, ROAS jumped to 2,308% and more page-view conversions surpassed 50,000.
“For CTV campaigns, Roku has been a top performer,” said Claire Folkestad, Paid Media Strategist, Blu Dot. “Comping to our other platforms, we have seen really strong ROAS… and highly efficient CPMs, lower than any other CTV partner we've worked with.”
Using Roku Ads Manager, the campaign moved from a pilot to a permanent performance engine for the brand.
Meme of the day
