New course: Transformers in Practice. You'll get a practical view of how transformer-based LLMs work, so you can reason about their behavior, diagnose problems like slow inference, and make smarter decisions about deployment. This course is built in partnership with @AMD and taught by @realSharonZhou. You'll see how transformers generate text one token at a time, how the model decides which earlier words matter most when predicting the next one, and how techniques like quantization speed up inference on GPUs. This is not a video-only course; interactive visualizations throughout let you play with these concepts and build intuition that sticks. Skills you'll gain: - Understand why LLMs hallucinate, and RAG and chain-of-thought shape what they generate - Look inside the model to see how attention and layers combine to predict the next token - Diagnose inference bottlenecks and learn the techniques that speed up transformers on GPUs Join and understand what's really happening inside your LLMs:

Meet China’s real-life Transformers. These powerful rescue platforms deployed to SW China’s Guangxi provide vital lifelines in raging floods and carry people out of danger.

要理解 LLM，必须读的论文顺序： 1. Attention Is All You Need（Transformers） 2. GPT-2（Scaling + Zero-shot） 3. Scaling Laws（Kaplan, 2020） 4. GPT-3（Few-shot） 5. Chinchilla（Data Needed） 6. InstructGPT（RLHF） 7. LoRA（Fine-Tuning w/o Broken） 8. FlashAttention（Fast） 9. Chain-of-Thought（Reasoning） 10. DPO（RLHF w/o Pain）

🚀 vLLM-Omni v0.20.0 is out — aligned with upstream vLLM v0.20.0 (CUDA 13.0 · PyTorch 2.11 · Transformers 5.x). ⚡ Qwen3-Omni throughput +72% on H20, 32 conc (0.241 → 0.414 req/s) via talker / code2wav multi-replica scaling 🎙️ TTS faster & leaner: VoxCPM2 RTF 0.946 → 0.106 · Fish Speech Fast AR latency -53% · Qwen3-TTS / Voxtral-TTS Code2Wav saves ~3.2 GiB 🎨 Diffusion dynamic step-level batching: +7.8% throughput / -5.8% latency 🆕 New / improved: HunyuanImage-3.0, ERNIE T2I, AudioX, Wan2.2-S2V, LTX-2.3, FastGen Wan 2.1 📱 Wan2.2 on NPU production-ready: MindIE-SD, fused ops, VAE BF16, HSDP/USP — +50–60% perf 🧮 Quant expanded: Qwen Omni W4A16, OmniGen2 FP8, Z-Image FP8, HunyuanImage3 NPU, GLM-Image 🧩 Multi-backend updates across CUDA / ROCm / MUSA / NPU / XPU Check it out →

本周三篇： Value Leakage: An LLM's Answers Are Silently Shaped by Its Own Values Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers Video = World + Event Stream

I don’t care what people say, I love Shia LaBeouf. First, he’s an incredible actor who often gets overlooked because people mostly remember him from movies like Transformers. But I’ll never forget seeing him in Fury for the first time and genuinely thinking he was outshining Brad Pitt. Second, I’ve watched a lot of his interviews, and you can see the struggle in him. You can also see someone who keeps trying to understand himself, become better, and move forward. I genuinely believe there’s a good person in there, and I hope he eventually finds peace and happiness.