Super excited to announce seven new world-class MAI models today. They represent what we consider a new era in AI designed to keep you in control and on the frontier. First is our text foundation model, MAI-Thinking-1, exceptionally strong on reasoning and SWE tasks. - It’s a 35B active parameter MoE with a 256K context window. Independent human raters on Surge prefer it for overall quality in blind side-by-sides versus Sonnet 4.6, and it’s achieved 97% on AIME 2025, the key measure of its general-purpose reasoning abilities. - It's at 53% on SWE Bench Pro, placing it right alongside Opus 4.6 on one of the toughest coding benchmarks. - And since we co-designed our models with our own silicon, MAI-Thinking-1 is optimized on our MAIA 200 chip. Benchmarking head-to-head against the GB200, we see 30% better performance per dollar as well as a 1.4x performance-per-watt gain when running our MAI models on the MAIA 200 end-to-end. Next is MAI-Image-2.5 and its Flash variant. Two super strong models now at #2# on the leaderboards, surpassing the score of Nano Banana 2 on image editing. Last for now is MAI-Code-1-Flash, our new inference efficient coding model, especially tuned for VS Code and GitHub Copilot CLI. - Code-1-Flash achieves 51% on SWE Bench Pro, despite having just 5B parameters, putting it closer to Haiku in size but cheaper in cost. All of this is the foundation for Microsoft Frontier Tuning. It lets you customize our models to create custom, company-specific agents that only you control. You can make our model, your model. Your data. Your agents. Your moat. Early adopters are already seeing a difference. When we tuned our models for McKinsey’s tasks, MAI delivered the highest win rate, outperforming GPT-5.5 on quality, while being 10x lower on cost. Also really excited to be collaborating with the amazing team at Mayo Clinic to jointly train a new frontier AI model for healthcare. Our announcements today mark another milestone on the road to humanist superintelligence. You can learn more and about our other new models in our latest blog:

Pipeline growth is getting harder. Sales teams are expected to reach more prospects, and qualify leads faster — yet most outbound still relies on manual dialing, limited bandwidth, and burned-out SDRs. Agora’s 𝗖𝗼𝗻𝘃𝗲𝗿𝘀𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁𝘀 change that.

The Chinese Communist Party’s state television @CCTV really has a "remarkable" talent for staging reality. A netizen in Sichuan noticed that the river near his home had turned muddy yellow after heavy rain. But on the live broadcast from CCTV, the same river appeared crystal clear and even bluish. Do you know how they managed to do that?

Today, we’re announcing Viktor’s $75M Series A, led by @Accel . @viktor__com was supposed to be a small experiment. It became the AI coworker 10x'ing real businesses. $15M in annualized revenue run rate. In 10 weeks. – Small companies saving millions of dollars – Sourcing hundreds of thousands in new revenue in their first 30 days – Whole teams getting half their week back – Companies running 40% leaner without cutting output Viktor is not another AI tool. It’s the first true AI employee. The vision that has been with us since 2023 when we started the company has finally been shipped. Back then, it was just the two of us, with a very small but dedicated team, iterating for years. Failing multiple times. Showing products that users didn't even want to test! But we never gave up. Our decisions were often wrong. Certainly more often than not! We kept trying. Now we’ve shipped something people love. Worth every sleepless night. Every sacrifice. The best employees don’t need to be told what to do. Neither does Viktor. Grateful to @Accel, our team, our earliest users, and everyone who believed this category could be bigger than chat.