kingmpw2020 @kingmpw2020

@GeminiApp Here comes, the SUNO killer... 🫡

It's Chinese New Year and I'm about to leave home. So I asked @GeminiApp to compose a farewell track — folk-pop, shifting from festive warmth to quiet solitude. Seconds later, Lyria 3 delivered a 30-second piece that genuinely moved me. 30 seconds is short. But when AI captures emotion this well, it hits different. 🎵

Introducing Lyria 3, our latest and most advanced music model, available in the Gemini App starting today : ) Go from idea, image, or video to music in seconds!

🎙️Ming-omni-tts-16.8B-A3B, Ming-omni-tts-0.5B, are the Voice Core of Ming-flash-omni-2.0, ready for: -Short-video creator looking for the perfect voiceover -Podcaster turns a monologe into a dynamic duo -Dev looking to integrate a voice assistant into your OpenClaw project

test

Intelligent Voice Design: Features 100+ premium built-in voices and supports zero-shot voice design through natural language descriptions

Fine-grained Vocal Control: Enables precise control over speech rate, pitch, volume, emotion, and dialects via simple instructions

keyboard_arrow_left Previous keyboard_arrow_right Next

One for all, and all for one 🧧 Introducing Ming-flash-omni-2.0: A specialist in every domain, unified as a capable generalist. A gift from Ling =) - Unified Acoustic Synthesis: Speech, audio, and music combined for unbounded creativity; - "Seeing" to "Knowing": Moving beyond input to true deep semantic understanding; - Native Visual Fusion: Seamless generation, editing, and segmentation;

推荐一个值得看 5678遍的五星级知识视频： 《光刻机背后的疯狂工程学》（The Ridiculous Engineering Of The World's Most Important Machine） 10 小时前刚发布的，55 分钟长度，已经有 200 万次观看。 我们家每天早饭时都会看英文的知识视频，veritasium 是必看（之前的 kurzgesagt、ted-ed 啥的都基本看完了）。 我5 年前读过asml 的传记《光刻巨人》，准备借此机会推荐给家里小学生也读一读，哈哈 整理了相关资料，按需自取： 1 视频链接：youtube.com/watch?v=MiUHjL… 2 notebooklm 做的 PPT（图 2-4）：img.candobear.com/20260101100700… 3 我还在 notebooklm 里面做了一个笔记本，设置成共享了，大家可以体验一下： notebooklm.google.com/notebook/ed811…

keyboard_arrow_left Previous keyboard_arrow_right Next

Whenever I feel lost, I watch this.

(2/3) Context-Aware ASR and Superior Dialect Recognition: The model achieves full SOTA across all 12 ContextASR subtasks. Its ability to understand and process 15 distinct Chinese dialects (e.g., Hunanese) is substantially improved. This advanced audio comprehension offers critical real-time translation and interpretation support, aiding users navigating unfamiliar linguistic contexts. #LLM #SpeechRecognition 🌍

(4/4) Enhanced Voice Cloning: The voice generation capability is significantly upgraded through a transition from discrete to continuous tokenizers. This advancement dramatically improves voice cloning accuracy and ensures high stability for both Mandarin and English mixed-language dialogue. The model can effectively clone the original speaker's timbre into new dialogue, achieving a remarkable seed-tts-zh WER score of 0.99, surpassing Qwen3 Omni and Seed-TTS. #VoiceCloning

@AntLingAGI awesome

🚀We officially release Ring-1T, the open-source trillion-parameter thinking model built on the Ling 2.0 architecture. Ring-1T achieves silver-level IMO reasoning through pure natural language reasoning. → 1 T total / 50 B active params · 128 K context window → Reinforced by Icepop RL + ASystem (Trillion-Scale RL Engine) → Open-source SOTA in natural language reasoning — AIME 25 / HMMT 25 / ARC-AGI-1 / CodeForce Deep thinking · Open weights · FP8 version available We're still training and pushing to do more. Stay tuned! 1/5

🚀 Ling-1T — Trillion-Scale Efficient Reasoner Introducing Ling-1T, the first flagship non-thinking model in the Ling 2.0 series — 1 Trillion total parameters with ≈ 50 B active per token, trained on 20 T+ reasoning-dense tokens. Highlights → Evo-CoT curriculum + Linguistics-Unit RL for scalable reasoning → Strong efficiency–accuracy balance on complex reasoning tasks → Advanced visual understanding + front-end code generation via Syntax–Function–Aesthetics reward → Emergent tool-use ability (≈ 70 %) with minimal instruction tuning → FP8 mixed-precision + Ling Scaling Law → efficient trillion-scale training Efficient Thinking · Precise Reasoning Ling-1T extends the Pareto frontier of reasoning accuracy vs. cost — a new milestone in open-source trillion-scale intelligence.

🫡 every week we treat your fomo with open AI release highlights: > new Qwen3-VL models (3B/30B MoE) better than GPT-5-Mini 🔥 > two new DeepSeek models (v3.2 experimental) > Ovi is a new text-to-video model with sounds ⏯️ > new Ming image and audio editing models find more ⤵️

今年最重磅的播客之一，警句汹涌而出 Lex Fridman与 Telegram 创始人 Pavel Durov（ @durov )长达 4 个多小时的对话： “这是我一生中最精彩、最有影响力的对话之一” 两人探讨了方方面面，从Durov的自由哲学到政府官僚机构、情报机构、人性、数学、加密、伟大的工程与设计、教育、家庭，以及他的人生哲学。 Durov 每天坚持做 300 个俯卧撑和 300 个深蹲...... 不是为了改变身体或增加肌肉，而是为了严格自律， 这是采访中最大的金句之一 时间戳： 0:00 - 简介 3:07 - 自由哲学 6:15 - 禁酒 14:20 - 禁手机 20:16 - 纪律 41:28 - Telegram：精益哲学、隐私和地缘政治 56:50 - 在法国被捕 1:13:01 - 罗马尼亚大选 1:23:56 - 权力与腐败 1:33:29 - 强化教育 1:45:29 - 尼古拉·杜罗夫 1:49:58 - 编程与电子游戏 1:54:11 - VK 的起源与工程 2:11:24 - 聘请优秀团队 2:20:40 - Telegram 工程与设计 2:39:42 - 加密 2:44:39 - 开源 2:49:26 - 爱德华·斯诺登 2:51:58 - 情报机构 2:53:10 - 伊朗以及俄罗斯政府的压力 2:56:19 - 苹果 3:03:16 - 中毒 3:29:28 - 埃隆·马斯克 3:35:31 - 金钱 3:44:23 - TON 3:54:13 - 比特币 3:57:12 - 两把椅子的困境 4:03:52 - 孩子们 4:15:02 - 父亲 4:19:33 - 量子永生 4:26:05 - 卡夫卡

Lex Fridman @lexfridman

12 months ago

Here's my 4+ hour conversation with Pavel Durov (@durov), founder and CEO of Telegram. This was one of the most fascinating and powerful conversations I've ever had in my life. We discuss everything from his philosophy on freedom to government bureaucracies, intelligence

726 2K 10K 2.6M 6K

🚀 Introducing Ming‑UniVision & MingTok — the first autoregressive model to natively unify vision understanding & generation in a continuous unified representation space. First unified continuous visual tokenizer . >3.5x faster convergence Seamless multi‑round interaction Visualized Chain-of-Thought (CoT)