Igor
160 posts
That's us! 🌍 The Artemis II crew captured beautiful, high-resolution images of our home planet during their journey to the Moon. As @Astro_Christina put it: "You guys look great."
We’re saying goodbye to the Sora app. To everyone who created with Sora, shared it, and built community around it: thank you. What you made with Sora mattered, and we know this news is disappointing. We’ll share more soon, including timelines for the app and API and details on preserving your work. – The Sora Team
Here's the answer: Breastfeeding! The oldest participants in the U.K. Biobank tend to report being breastfed as kids, whereas for the youngest, it's a coin flip.
Inworld TTS-1.5 Max just debuted as #1 on the Artificial Analysis TTS leaderboard, with our TTS-1 Max at #2. These are third-party evaluations based on blind user tests. TTS-1.5 achieves this through 30% greater expressiveness and 40% fewer word errors compared to TTS-1, which was already the top model. Our gap against other models is expanding. TTS-1.5 also delivers 4x faster performance with sub-250ms P90 latency and 25x lower cost than alternatives at under 1 cent per minute of interaction. We'll be publishing more detailed evaluations and competitive benchmarks on these metrics soon.
Yann LeCun says the AI industry is completely LLM-pilled, with everyone digging in the same direction and stealing each other's engineers "i left Meta because they also became LLM-pilled" We cannot build true agentic systems without the ability to predict the consequences of actions, just like humans do
Qwen3-TTS is officially live. We’ve open-sourced the full family—VoiceDesign, CustomVoice, and Base—bringing high quality to the open community. - 5 models (0.6B & 1.8B) - Free-form voice design & cloning - Support for 10 languages - SOTA 12Hz tokenizer for high compression - Full fine-tuning support - SOTA performance We believe this is arguably the most disruptive release in open-source TTS yet. Go ahead, break it and build something cool. 🚀 Everything is out now—weights, code, and paper. Enjoy. 🧵 Github: github.com/QwenLM/Qwen3-T… Hugging Face: huggingface.co/collections/Qw… ModelScope: modelscope.cn/collections/Qw… Blog: qwen.ai/blog?id=qwen3t… Paper: github.com/QwenLM/Qwen3-T… Hugging Face Demo: huggingface.co/spaces/Qwen/Qw… ModelScope Demo: modelscope.cn/studios/Qwen/Q… API: alibabacloud.com/help/en/model-…
Inworld TTS-1.5 releases today. The #1 TTS on Artificial Analysis now offers realtime latency under 250ms and optimized expression and stability for user engagement, and costs half a cent per minute. Some voice models are fast, some are expressive, some are affordable. We outperform them all across the board. Production-grade realtime latency: <250ms latency for Max model, <130ms for Mini (P90 first audio) - 4x faster than before. Voice agents now respond before users notice any delay. Engagement-optimized quality: 30% more expressive to serve a wider range of personalities and 40% lower word error rates for fewer hallucinations, word cutoffs, and audio artifacts. Built for consumer-scale: Radically affordable with enhanced multilingual support (15 languages including Hindi) and enhanced voice cloning, now via API. On-prem options now available for enterprises.
Our CEO @KylanGibbs met with Jensen Huang to discuss consumer-grade AI and what it takes to power the next wave of applications. The first wave of AI made existing things cheaper: automation, efficiency, cost savings. The next wave makes things that couldn't exist before. New products, new experiences, new revenue. AI reaches its real economic potential when it creates value consumers want to pay for, not just value businesses want to save. What consumer-grade AI requires: - Realtime: Sub-300ms response for conversations that feel instant - Scalable: Reliable infrastructure for millions of concurrent users, not enterprise pilots - Personal: Engaging, expressive, adapted to each user through constant experimentation That's what it takes to bring AI to everyday life.
