MiMo-V2.6 Agentic RL: How Xiaomi Scaled Reinforcement Learning Across Billions of Tokens
19 days ago 高效码农MiMo-V2.6 Technical Report, Explained: What Scaling Agentic Reinforcement Learning Actually Takes Xiaomi’s LLM-Core team set out to answer a narrow question in this report: if you scale reinforcement learning along three axes at once, compute, environment diversity, and grader compute, does model capability keep climbing? The answer is yes. DeepSWE jumped from 19.0 in the previous generation to 71.9, CyberGym from 40.0 to 94.0, and the team open-sourced the training dynamics, RL environments, and end-to-end framework behind those numbers. The report runs over forty pages and covers architecture, pre-training, mid-training, RL, infrastructure, and open reproduction. This walkthrough reorders the material …
I Ran a 2.78 Trillion Parameter Kimi K3 LLM on 8GB RAM with No GPU: Here’s How It Actually Works
1 months ago 高效码农I Ran a 2.78 Trillion Parameter LLM on 8GB RAM with No GPU: Kimi K3 in C The short version, upfront: This isn’t marketing fluff about “running a 2.78T model in 8GB.” I actually ran the inference engine on a standard laptop with only 8 GB of RAM, got the correct first token, and verified that it produces byte-for-byte identical outputs compared to a 224 GB server rack. The catch? Each token takes about half a minute. This post is a hands-on log of setting it up—downloading the 1.56 TB checkpoint, packing the trunk, running the same prompts across different …
AntAngelMed: How to Deploy the World-Leading Open-Source Medical LLM in Your Hospital
9 months ago 高效码农Bringing the “Hospital Brain” Home: A Complete, Plain-English Guide to AntAngelMed, the World-Leading Open-Source Medical LLM Keywords: AntAngelMed, open-source medical LLM, HealthBench, MedAIBench, local deployment, vLLM, SGLang, Ascend 910B, FP8 quantization, 128 K context 1. What Is AntAngelMed—in One Sentence? AntAngelMed is a 100-billion-parameter open-source language model that only “wakes up” 6.1 billion parameters at a time, yet it outscores models four times its active size on medical exams, and you can download it for free today. 2. Why Should Non-PhD Readers Care? If you code: you can add a medical “co-pilot” to your app in one afternoon. If you …
Causal-Attention Diffusion LM: How WeDLM Outperforms vLLM Without Custom Kernels
9 months ago 高效码农WeDLM in Practice: How to Deploy a Causal-Attention Diffusion LM That Outruns vLLM Without New Kernels TL;DR: WeDLM keeps causal attention, reorders tokens so masked positions still see all observed context, and commits tokens left-to-right as soon as they are predicted. The result is the first diffusion-style language model that beats a production vLLM baseline in wall-clock time while preserving (and sometimes improving) accuracy. This post explains why it works, how to run it, and what to watch when you ship it. What exact problem does WeDLM solve? Question answered: “Why do most diffusion language models feel fast in papers …
Qwen3-8B-Drama-Thinking: How AI Screenwriting Reveals Its Creative Process
10 months ago 高效码农Qwen3-8B-Drama-Thinking: When AI Starts “Thinking” About Screenwriting Core question: How does this model elevate AI scriptwriting from text generation to demonstrating creative thinking? Qwen3-8B-Drama-Thinking is an 8-billion parameter large language model specifically designed for screenwriting. Its breakthrough lies not in producing better scripts, but in visualizing the entire creative process on screen—wrapping three to four thousand tokens of reasoning chains within <think>…</think> tags that meticulously detail everything from thematic deconstruction and character psychology analysis to three-act structure planning. This isn’t mere text generation; it’s a “visualization” of the creative workflow. 1. Core Features: Why It’s a “Creative Thinking Partner” Central …