#SO architecture
5236 related articles
·3 min
Deep Dive into the Kimi-K3 Technical Report: Architecture, Long Context, and Reasoning Capabilities
Deep analysis of Moonshot AI's Kimi-K3 technical report covering long context processing, MoE architecture, reasoning capabilities, and China's position in the global AI competition.
Read more →
·3 min
Deep Dive into the Kimi-K3 Technical Report: A Comprehensive Analysis of Architecture, Long-Context Processing, and Reasoning Capabilities
Deep dive into Moonshot AI's Kimi-K3 technical report, analyzing its long-context processing, MoE architecture, reasoning improvements, and its position in global AI competition.
Read more →
·4 min
Building Self-Hosted Tunnels Under Extreme Censorship: A Layered Architecture with VLESS Reality
Deep analysis of building layered tunnel architectures using VLESS+Reality to evade DPI and domestic relays with whitelisted IPs to bypass geographic blocking in extreme censorship environments.
Read more →
·4 min
Vibe Coding in Practice: AI-Assisted Refactoring of a Unity Game's SO Architecture
Learn how to use Vibe Coding to collaborate with AI in refactoring a Unity game, introducing ScriptableObject (SO) data architecture, with hands-on tips.
Read more →
·4 min
Vibe Coding in Practice: Refactoring a Unity Game's SO Architecture with AI
A deep dive into refactoring a Unity game with Vibe Coding and AI, introducing ScriptableObject (SO) architecture—covering visual map editors, Codex remote control, and collaboration pitfalls.
Read more →
·3 min
Rocky: An Open-Source Coding Agent with Minimal Architecture and Native DeepSeek Integration
Rocky is a minimal, transparent open-source coding agent with a core loop under a few hundred lines of Python, native DeepSeek search, Research mode, and built-in SWE benchmarking for reproducible agent experiments.
Read more →
·2 min
Voice AI Agent Architecture: Pipecat Flows vs Vapi Squad — A Deep Comparison
A deep comparison of Pipecat Flows and Vapi Squad for voice AI agent architecture — covering latency, accuracy, multi-agent handoffs, and when to use each.
Read more →
GitHub Daily · July 22: Financial Foun…
·3 min
GitHub Daily · July 22: Financial Foundation Model Kronos Leads, Architecture Visualization and the Claude Skills Ecosystem Soar
GitHub Trending July 22: Kronos brings the foundation model paradigm to finance, awesome-claude-skills tops with 70K stars, plus LikeC4 architecture-as-code and Rust-based Pumpkin.
Read more →
·3 min
Qwen's Six-Generation Evolution: A Complete Breakdown from Architecture Innovations to the Reasoning Race
From Qwen1 to Qwen3-2507: a complete breakdown of six generations of evolution over two years — GQA, MoE, GRPO, dynamic thinking, and the core shift from architecture to reasoning.
Read more →
·2 min
GPT-Live Full-Duplex Voice Model: Listen While Speaking + Reasoning Delegation Architecture Explained
OpenAI's GPT-Live full-duplex voice model enables natural simultaneous conversation with a reasoning delegation architecture pairing real-time dialogue with GPT-5.5 deep reasoning. Now live for 150M users.
Read more →
·2 min
HF Viewer: Free Interactive Visualization Tool for Exploring 2,300+ Open-Source AI Model Architectures
HF Viewer is a free interactive tool for visualizing 2,300+ open-source AI model architectures. Explore Transformers and more via graph nodes, animations, and paper links.
Read more →
·2 min
ComfyUI v0.28.0: Native Video Super-Resolution, 3D Support, and PixelDiT Architecture — A Complete Breakdown
ComfyUI v0.28.0 adds SeedVR2 native video super-resolution, PixelDiT architecture, 3D Gaussian Splatting export, int4 quantization, and lip-sync integration for a major multimodal AI workflow upgrade.
Read more →
·3 min
Can Small Local Models Become Software Domain Experts? A RAG Architecture Deep Dive
Can small local models (1.5B–3B) become software domain experts? This article breaks down CPT, SFT, RAG, and Agent architectures, with a layered RAG-centric design for CPU-only local deployment.
Read more →
Agentic Loop Explained: The Three-Loop…
·2 min
Agentic Loop Explained: The Three-Loop Architecture of Reasoning, Tool Use, and Orchestration
A deep dive into the Agentic Loop — breaking down the three-layer architecture of reasoning, tool use, and orchestration to help developers build and debug reliable AI agent systems.
Read more →
·2 min
When Does Multi-Agent Architecture Slow Down Your Workflow? A Single vs. Multi-Agent Comparison
Multi-agent architecture isn't always better. This article analyzes how splitting tasks across agents introduces latency, token costs, and error propagation — with a practical framework for deciding when to use single vs. multi-agent design.
Read more →
Hardware-Software Co-Design: A Guide t…
·2 min
Hardware-Software Co-Design: A Guide to Building Hardware-Friendly LLM Architectures
Explore AI Model Co-Design principles and how hardware-friendly LLM architecture design — covering MoE, GQA, and FP8 quantization — optimizes the accuracy, throughput, and latency trade-off.
Read more →
Dense: An Open-Source ML Workbench Bui…
·2 min
Dense: An Open-Source ML Workbench Built for Neural Network Architecture Research
Dense is an open-source ML IDE for neural network architecture research. It integrates the DeltaImportance layer and architecture visualization to help researchers iterate faster and analyze network importance during the design phase.
Read more →
·4 min
Gemini Accidentally Leaks Its Internal Reasoning Architecture: A Deep Dive into Bento Components and Knowledge Graph
An accidental prompt leak revealed the inner workings of Google Gemini's reasoning and UI rendering architecture, including Bento card components, the chameleon adaptive system, and knowledge graph entity ID retrieval.
Read more →
·4 min
AI Persistent Memory Three-Layer Architecture Explained: In-Depth Comparison of Mem0, Zep, and ContextNest
Deep dive into the three-layer architecture of AI persistent memory systems—storage, management, and retrieval—with an in-depth comparison of Mem0, Zep, and ContextNest to help developers choose the right memory solution for AI Agents.
Read more →
·2 min
Tencent Hunyuan V3 Open-Sourced: Full Breakdown of the 295B MoE Architecture — 256K Context Is Its Biggest Weakness
Tencent open-sources Hunyuan V3: 295B MoE model with 21B active params, claims to beat DeepSeek-V4-Pro. Supports FP8, vLLM/SGLang — but 256K context and no multimodal are key drawbacks.
Read more →