Voice AI has an evaluation problem. Models look strong... on public benchmarks, then collapse on real-world audio. Introducing a recipe-driven evaluation framework for low-resource languages, real-world audio, and production failure modes. Details ↓show more

🔥Spatial intelligence requires world generation, and now we have... the first comprehensive evaluation benchmark📏 for it! Introducing WorldScore: Unifying evaluation for 3D, 4D, and video models on world generation! 🧵1/7 Web: arxiv:show more

AI was trained on the open internet, but the... data that matters most lives in the real world. Introducing early access to Numo, an app built to collect the next generation of AI training data. Starting with voice data collection in Bengali, Hindi, Tamil, and Telugu. Details ↴show more

⚡ Launching as a Day-0 partner: Tencent Hy Hy3... is now available on OpenRouter. 🎉 Free through July 21. Built for real-world business scenarios, Hy3 delivers: 🧠 295B MoE (21B active) 📚 Native 256K context ⚙️ Three configurable reasoning modes 💻 Strong coding, reasoning & long-context capabilities 🤖 Optimized for agentic workflows Designed for productivity and real-world AI workflows, balancing reliability, efficiency, and cost at production scale. #Hy3 #LLM #Agentsshow more

Introducing $SAHARA — the native utility token of Sahara... AI. Designed for real-world function, $SAHARA powers a new economic model for AI. One that is collaborative, open, and built to reward real contributors. Utility, distribution, and claim details below 🧵👇show more

Announcing Bespoke-MiniChart-7B, a new SOTA in chart understanding for... models of comparable size on seven benchmarks, on par with Gemini-1.5-Pro and Claude-3.5! 🚀 Beyond its real-world applications, chart understanding is a good challenging problem for VLMs, since it requires both mathematical as well as visual reasoning. 1/n🧵show more

How can robots reliably place objects in diverse real-world... tasks? 🤖🔍 Placement is tough—objects vary in shape and placement modes (such as stacking, hanging, and insertion), making it a challenging problem. We introduce AnyPlace, a two-stage method trained purely on synthetic data to predict diverse placement poses of unseen objects for real-world tasks. Read on for more👇show more

Google dropped another banger! They just released a comprehensive... white-paper on AgentOps - the missing piece between building AI agents and actually shipping them to production. Here's the reality: Building an AI agent takes minutes. Making it production-ready? That's where 80% of the real work begins. Google's "Prototype to Production" guide tackles this exact problem. The framework has three core pillars: 1. Evaluation-Gated Deployment: No agent reaches users without passing tests. Build a "golden dataset" that validates behavior, not just functionality. This catches what unit tests miss - agents choosing wrong tools or hallucinating responses. 2. Automated CI/CD for Agents: Test in stages: pre-merge checks for fast feedback, staging for load testing, then gated production. Version everything: prompts, tools, configs, evaluation datasets. 3. Observe → Act → Evolve Loop Production isn't the finish line. Monitor through logs, traces, and metrics. Act with circuit breakers and human escalation. Evolve by turning production failures into test cases. The best part? They released the Agent Starter Pack - a template with CI/CD, Terraform deployment, and built-in observability. Helps you spin up an evaluation pipeline in minutes. The guide also talks about the two major protocols and how they can work together. ↳ MCP for tool integration ↳ A2A for agent collaboration If you're shipping agents to production, you should read this. I've shared the full white-paper in the next tweet!show more

4/ Finally, I think future iterations of models like... Genie 3 will have a significant impact on accelerating robotics and real-world AI. Here's a glimpse of what that could look like: an agent pursuing a goal (go to tomatoes) in an environment generated by our model.show more

💻 Agent mode just went local! Today, Mistral AI... released Devstral, an Apache-2.0, 24B-parameter model trained for tool calling in real-world software development environments It reached #1 on SWE-bench for open-source models, but more importantly in our real-world testing it has proven to be capable of navigating and autonomously editing codebases, while running entirely on a laptop To try it out, just add `ollama/devstral` in Continue. See below for a quickstart with ollama👇show more

it's open source time, with a real leap for... world models 🎉 NVIDIA's SANA-WM: a camera-conditioned world model that fits on one GPU. 60s of 720p in 34s on a single 5090 - 2.6B params and Apache 2.0!show more

IoTeX was selected by Messari as a core case... study in its new AI report, cited alongside World, Coinbase 🛡️, Openτensor Foundaτion and others. Messari’s research underscores a bigger shift: real-world AI needs verified data, and IoTeX is the first full stack infrastructure that can supply it. Live device data becomes verifiable insight, ready for real-world agentic deployment. Quicksilver is already proving this in production with DIMO, Nubila Network and more, powering agentic applications built on IoTeX-verified real-world intelligence. Read the full analysis in the report below.show more

Grok Voice Agent API is now one of the... most advanced voice AIs in the world It ranks #1 on BigBench Audio for speech reasoning Grok Voice already powers voice mode across Grok apps and runs in millions of Tesla vehicles Conversations feel so natural, expressive, fluid, and instantly responsive - like talking to a real person, not a machine Try it now on Grok, in Tesla, and across the growing ecosystem of apps and experiences powered by xAIshow more

Excited to share a few presentations, demos, and workshop... talks from our group and collaborators at #ICRA2026! We will present recent work on real-to-sim-to-real robot policy evaluation, model-based planning with learned dynamics, and multi-modal manipulation. We will also have a joint live demo between SceniX and Analog Devices, Inc. on real-to-sim-to-real cable manipulation at the ICRA exhibition. This is a small teaser of what we have been building, with more to come soon! If you are at ICRA, please stop by the sessions or the demo booth. Happy to chat about robot learning, simulation, world models, and sim-to-real!show more

Introducing a Major Update on AIOZ AI V1: AI... Challenges Are Here! AIOZ AI Challenges is a collaborative space where AI builders and tech enthusiasts can create, submit, and explore innovative models tackling real-world problems. This isn’t just a leaderboard — it’s a launchpad for hands-on learning, community collaboration, and potential rewarded innovation. Challenge One: "Face Anti-Spoofing" Starts Now This challenge dives deep into advanced computer vision—tasking you to detect spoofing attacks like printed photos, video replays, and 3D masks, which are critical threats to real-world biometric systems. Rewards: Upload your model to AIOZ AI V1 and unlock a chance to receive a share of the 4,500 AIOZ token challenge rewards. Full Challenge Details & Submission Guidelines: Build your model on AIOZ AI—the Web3 collaborative marketplace powering the DePIN Ecosystem for AI.show more

Most AI models just predict the next pixel to... build a narrative or an image. Gemini Omni is different. Gemini Omni has an intuitive understanding of forces like gravity, kinetic energy, and fluid dynamics for more realistic movements that follow real-world logic.show more

If you’re designing an app with audio, get ready... for the most feature-rich audio support coming in our next release! Some highlights: › Upload audio files and access the microphone for input › Generate real-time waveform data based on input (audio or mic) › currentTime(), TotalTime(), volume, pan, rate, numberOfLoops supported for interactions › Sync multi tracks This release will have mic support and the speech framework will be supported in the future. This project and others will be available in the gallery for you to duplicate and make your own!show more

so, i might have been wrong.... i've been looking... into AI UGC a little bit more and it's actually kind of insane the video you see below took $2 to make and the only visible problem (might) be the audio i know ecom brands using ai ugc for there organic and paid doing over $1M/mo i'm still bullish on real human clipping and ai content but considering how much ai ugc has improved over the last few months... it's going to be interesting to see how the content marketing world changes this yearshow more

Most video models do one thing well. MiniMax H3... does everything, together. Our next-gen open-weight multimodal model reads text, image, video, and audio as a single creative language: motion, sound, emotion, cinematography, all connected. The result: precision-level generation with real cinematic quality. → Commercial-grade: film, ads, MVs, UI, game CG → 2K/24FPS, native stereo audio → Voice cloning + multi-asset reference for characters, motion & camera Concept to production. One seamless flow. | #MiniMaxH3 Hailuo AI (MiniMax)show more

OVER x Silencio We’re partnering with Silencio | Voice... Data for AI to add a new layer of data to the OVER mapping experience. Starting today, users can optionally capture real-world sound while creating their 3D maps inside the OVER app, and we seamlessly share the noise data with Silencio. This will soon allow mappers who capture both visuals and sound to also unlock additional rewards for their contribution. 🌎 We map physical locations around the world. 🌎 Silencio maps the world through sound. Together combining visual + audio signals , we’re enriching the foundation of Physical AI with deeper, multi-layered real-world data.show more

Big moment for text-to-speech. Qwen just open-sourced a text-to-speech... model that lets you clone voices, design new ones, and control speech using natural language. Let me explain what I mean: You can literally tell it "speak in a cheerful tone with slight nervousness," and it actually does that. No complex audio engineering needed. What makes this special: - 3-second voice cloning - Covers 10 languages: English, German, French, and more - Latency as low as 97ms for real-time applications - Supports both streaming and non-streaming generation The model comes in two sizes (0.6B and 1.7B parameters), so you can pick based on your hardware and quality needs. Three modes to work with: 1. Custom Voice: Use pre-built premium voices with instruction-based style control 2. Voice Design: Describe the voice you want in plain English (or Chinese), and the model creates it 3. Voice Clone: Provide a 3-second reference audio and clone that voice The best part? It integrates with vLLM for production deployment and has a simple Python package you can pip install. I've shared a link to the GitHub repo in the next tweet.show more