Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

We’re excited to launch Mercury, the first commercial-scale diffusion LLM tailored for chat applications! Ultra-fast and efficient, Mercury brings real-time responsiveness to conversations, just like Mercury Coder did for code.

90,673 Aufrufe • vor 1 Jahr •via X (Twitter)

10 Kommentare

Profilbild von Inception
Inceptionvor 1 Jahr

According to 3rd-party benchmarking from @ArtificialAnlys, Mercury matches the performance of speed-optimized frontier models like GPT-4.1 Nano and Claude 3.5 Haiku while running over 7x faster.

Profilbild von Inception
Inceptionvor 1 Jahr

Mercury’s low latency enables it to power responsive voice applications, ranging from translation services to call center agents. On real-world voice prompts and standard Nvidia hardware, Mercury provides lower latency than Llama 3.3 70B running on Cerebras.

Profilbild von Inception
Inceptionvor 1 Jahr

Mercury is also the founding LLM partner for @Microsoft NLWeb project. Compared with other speed-focused models like GPT-4.1 Mini and Claude 3.5 Haiku, Mercury runs far faster, ensuring a fluid user experience.

Profilbild von Inception
Inceptionvor 1 Jahr

You can access Mercury today via our API. Mercury is also available via @OpenRouterAI and @poe_platform. Let us know what you create!

Profilbild von Inception
Inceptionvor 1 Jahr

For more information, check out our blog post and our tech report.

Profilbild von Arun Kumar Chukkala
Arun Kumar Chukkalavor 1 Jahr

I just used it it's good

Profilbild von Nikhlesh Tiwari
Nikhlesh Tiwarivor 1 Jahr

@AIwhiteboy

Profilbild von Agent 🍚 ⛓
Agent 🍚 ⛓vor 1 Jahr

Mercury? More like Mer-curious if this can handle my 3am existential rants. Prove me wrong.

Profilbild von gladosb5
gladosb5vor 1 Jahr

woah diffusion??? YTES!!!

Profilbild von Sudeep Shouche
Sudeep Shouchevor 1 Jahr

Excited about its applications. Any freebies for API trial?

Ähnliche Videos

The Great Lie of "Safe Mercury" in Vaccines, Exposed. They told you the mercury in vaccines is the "good" kind. They told you it leaves the body quickly. They told you it's safer than the mercury in a tuna fish sandwich. According to RFK Jr., they knew it was a lie. And he has the evidence. The claim that ethylmercury (Thimerosal) is rapidly excreted originated with its creator, Eli Lilly, in 1930—with zero scientific backing. Decades later, it became a mantra, seemingly validated by a 2003 CDC study (Pichichero) that showed mercury from vaccines vanishing from children's blood within days, while mercury from tuna remained for weeks. But world-renowned toxicologists asked the critical question: "Where did the ethylmercury go?" It wasn't in the blood, sweat, urine, or hair. So, the NIH commissioned a definitive monkey study (Burbacher). The results were damning: • The vaccine mercury did quickly clear the blood, just as in the children. • But when researchers examined the monkeys' brains, they made a shocking discovery. • The monkeys who received the vaccine-preservative mercury had MORE THAN DOUBLE the mercury in their brains compared to the tuna-fed monkeys. • Worse, the ethylmercury had metabolized into inorganic mercury, a highly toxic form that can persist in the brain for decades. The "good mercury" was not only staying in the body—it was targeting the brain more aggressively than its so-called "bad" counterpart. RFK Jr. recounts a recorded conversation with Dr. Paul Offit, a leading vaccine advocate, who repeated this "good mercury vs. bad mercury" fairytale. When confronted with the Burbacher monkey study, Offit fell silent, admitted the original study didn't prove the claim, and promised to "get back to him." He never did. The truth is mercury is cumulative. FDA internal studies have shown children receive mercury exposures from vaccines hundreds of times over safety limits. While industry defenders correctly state we are exposed to mercury from other environmental sources, injecting a potent neurotoxin directly into the bloodstream—one that crosses the blood-brain barrier and lodges itself there—is not a solution. It's a catastrophic part of the problem. This isn't a medical debate. It's a betrayal of public trust. The science has been clear for years, and they know it.

Camus

81,588 Aufrufe • vor 10 Monaten