LLMs (MoC) - DeveloPassion

# LLMs (MoC) ## Overview Notes about Large Language Models (LLMs). ## Notes   - [[2026-04-21 Kimi K2.6, Qwen, and Gemma 4 - Local AI Is Catching Up]] - [[2026-05-04 Heavy AI Agents Are an Anti-Pattern]] - [[2026-05-06 Gemma 4 Gets Multi-Token Prediction Drafters - 3x Faster Inference Without Quality Loss]] - [[2026-06-17 The US Government Banned Claude Fable 5 and Mythos 5]] - [[2026-08-01 Kimi K3, Qwen 3.8, and a Week That Reset the Open-Weight Frontier]] - [[2026-08-05 OpenAI shrank the Codex context window on purpose]] - [[2026-09-02 Claude Fable 5.1 - cheaper caching, calmer safeguards, same old Claudish]] - [[2026-09-15 See Which Tokens an LLM Draws From, Right in Your Browser]] - [[2026-09-22 Claude Opus 5.5 - Fable-level work for less than half the price]] - [[2026-09-22 Jev and System One models - a semantic if statement for your code]] - [[2026-09-29 Claude Sonnet 5.5 - Opus-level everyday work at half the price]] - [[Agent-Native Product Decomposition]] - [[AgentRun]] - [[AI Agent Routing]] - [[AI Assistant plugin for Obsidian]] - [[AI Assistants]] - [[AI Commander plugin for Obsidian]] - [[AI Engineering]] - [[AI Expert Offloading]] - [[AI Fine-Tuning]] - [[AI for Templater plugin for Obsidian]] - [[AI Frontier Model]] - [[AI Hallucination]] - [[AI Image Analyzer plugin for Obsidian]] - [[AI image generation models]] - [[AI Inference]] - [[AI Instruction Tuning]] - [[AI KV Cache]] - [[AI Major Techniques]] - [[AI Master Prompt]] - [[AI Mixture of Experts (MoE)]] - [[AI Model Calibration]] - [[AI Model Cascades]] - [[AI Multi-Token Prediction Drafters]] - [[AI Multimodal]] - [[AI Open Weight Models]] - [[AI Post-Training]] - [[AI Providers plugin for Obsidian]] - [[AI Quantization]] - [[AI Reasoning Models]] - [[AI Sampling Parameters]] - [[AI Scaling Laws]] - [[AI Sparse Attention]] - [[AI Speculative Decoding]] - [[AI Sycophancy]] - [[AI Tagger plugin for Obsidian]] - [[AI Temperature]] - [[AI Tokenization]] - [[AI Tool Use]] - [[AI Tools I use]] - [[AI Verifiability]] - [[AI Verifiability as a Capability Ceiling]] - [[AirLLM]] - [[Ambient Programming]] - [[Anthropic]] - [[Anthropic's position on open-weights models (2026)]] - [[Antigravity CLI]] - [[Apertus]] - [[Artificial Analysis]] - [[Auto Classifier plugin for Obsidian]] - [[Baichuan]] - [[Berget AI]] - [[Bonsai 27B]] - [[Browser MCP]] - [[Business AI Master Prompt Questions of Tiago Forte]] - [[Candidate-Then-Select Extraction]] - [[Cannoli plugin for Obsidian]] - [[Chain-of-Thought (CoT) prompting]] - [[Chat log to notes prompt]] - [[Chat Stream plugin for Obsidian]] - [[ChatGPT]] - [[ChatGPT Images 2.0]] - [[ChatGPT Images 2.5]] - [[Cheaper Inference]] - [[Claude]] - [[Claude 5]] - [[Claude Code]] - [[Claude Code effort levels]] - [[Claude Code Prompt Caching]] - [[Claude Fable 5]] - [[Claude Fable 5.1]] - [[Claude MCP Browser Extension]] - [[Claude Opus 4.7]] - [[Claude Opus 4.8]] - [[Claude Opus 4.8 and Dynamic Workflows]] - [[Claude Opus 5]] - [[Claude Opus 5.5]] - [[Claude Sonnet 5]] - [[Claude Sonnet 5.5]] - [[Claudian plugin for Obsidian]] - [[Cline]] - [[Code2Prompt]] - [[Codestral]] - [[CodeWeaver]] - [[Codex App]] - [[Codex CLI]] - [[Codex Cloud]] - [[Codex IDE Extension]] - [[Combining multiple LoRAs]] - [[ComfyUI]] - [[Companion plugin for Obsidian]] - [[Composer 2]] - [[Composer 2.5]] - [[Composite Scoring]] - [[Confidence-Gated Routing]] - [[Context Compression]] - [[Context Engineering]] - [[Context size reduction]] - [[Context Window]] - [[Context-Understanding-Generation Asymmetry]] - [[Context7]] - [[Copilot Notebooks]] - [[Copilot plugin for Obsidian]] - [[Cursor]] - [[Cursor Bridge plugin for Obsidian]] - [[DALL-E]] - [[Data Poisoning]] - [[Decision Models (DMs)]] - [[Deepseek]] - [[DeepSeek V3]] - [[DeepSeek v4]] - [[DeepSeek V4.1 Flash]] - [[Dense AI Models]] - [[Desktop Extensions for MCP]] - [[Devstral]] - [[Devstral 2]] - [[Dify]] - [[Docker Agent]] - [[DSLs Make LLM Output Reliable]] - [[DSPy]] - [[EmbeddingGemma]] - [[fal.ai]] - [[FastContext]] - [[files-to-prompt]] - [[FLUX.1]] - [[FLUX.2]] - [[FLUX.3]] - [[Gemini]] - [[Gemini 3]] - [[Gemini 3.1 Flash Live]] - [[Gemini 3.1 Flash TTS]] - [[Gemini 3.5 Flash]] - [[Gemini 3.5 Pro]] - [[Gemini 3.6 Flash]] - [[Gemini 4 Argon]] - [[Gemini App for Windows]] - [[Gemini CLI]] - [[Gemini Code Assist]] - [[Gemini Mobile App]] - [[Gemma]] - [[Gemma 4]] - [[Generative AI (Gen AI)]] - [[Generative AI Risks]] - [[GEPA]] - [[GitHub Copilot]] - [[GitHub Copilot plugin for Obsidian]] - [[Glama Chat]] - [[Google AI Studio]] - [[Google Cloud Knowledge Catalog]] - [[Google DeepMind]] - [[GPT Image 1]] - [[GPT-5]] - [[GPT-5.4]] - [[GPT-5.5]] - [[GPT-5.6]] - [[GPT-6 Astra]] - [[GPT-6 Luna]] - [[GPT-6 Sol]] - [[GPT-6.1 Sol]] - [[GPT4]] - [[Granite]] - [[Granite 4.1]] - [[Grok]] - [[Grok 3]] - [[Grok 4]] - [[Grok 4.3]] - [[Grok Build]] - [[Headroom]] - [[Heavy AI Agents Are an Anti-Pattern - Why Fewer Agents With More Skills Wins (Article)]] - [[Hermes]] - [[Hermes Agent]] - [[Hierarchical Classification]] - [[How I leverage my Notes with AI]] - [[How to combine multiple FLUX.1 LoRas]] - [[How to create a top notch AI writing assistant]] - [[How to create your Business AI Master Prompt]] - [[How to create your Personal AI Master Prompt]] - [[How to structure your AI Master Prompt]] - [[How to train a FLUX.1 LoRA]] - [[Imagen]] - [[Imgen Arena]] - [[InstructGPT]] - [[Jagged Intelligence]] - [[Jev]] - [[just-prompt MCP server]] - [[Kimi]] - [[Kimi K2.5]] - [[Kimi K2.6]] - [[Kimi K3]] - [[LangFlow]] - [[Large Language Models (LLMs)]] - [[LiteLLM]] - [[Living shim between humans and AI models]] - [[LLM Knowledge Bases Over Unstructured Data]] - [[LLM Monitoring]] - [[LM Studio]] - [[LM Studio Mobile App]] - [[Local GPT plugin for Obsidian]] - [[Loom plugin for Obsidian]] - [[Low Rank Adapter (LoRA)]] - [[Machine Native Intelligence]] - [[Magistral Medium]] - [[Magistral Small]] - [[Markdown-based Installation (MD Scripts)]] - [[Martian]] - [[MCP server for Obsidian (Python)]] - [[mcptools]] - [[Menugen Architecture Pattern]] - [[MenuGen Deployment Gap]] - [[Midjourney]] - [[Ministral 3]] - [[Miso One]] - [[Mistral Large 3]] - [[Mistral Medium]] - [[Mistral Medium 3.5]] - [[Mistral Small 3]] - [[Mistral Small 3.1]] - [[Mistral Small 4]] - [[Mode Collapse]] - [[Model Context Protocol (MCP)]] - [[Model routing]] - [[Moonshot AI]] - [[Nano Banana Pro]] - [[Not Diamond]] - [[Note Companion plugin for Obsidian]] - [[NotebookLM]] - [[Nous Chat]] - [[Nous Portal]] - [[NVIDIA API Catalog]] - [[Obliteratus]] - [[Odysseus (AI)]] - [[Odysseus Compare]] - [[Odysseus Cookbook]] - [[Ollama]] - [[OmniParser]] - [[Open Knowledge Format (OKF)]] - [[Open WebUI]] - [[OpenAI]] - [[OpenAI Codex]] - [[OpenAI Product Consolidation]] - [[OpenRouter]] - [[OpenRouter Fusion]] - [[openrouterai OpenRouter MCP Server]] - [[Orpheus TTS]] - [[Personal Knowledge Management is here to stay]] - [[Pixtral]] - [[Portkey]] - [[Pre-warm the Prompt Cache]] - [[Preparing for the future of knowledge work]] - [[Prompt Engineering Best Practices]] - [[Prompt Engineering Strategies]] - [[Prompt Lazy Loading AI Design Pattern (PLL)]] - [[Prompt-driven development (PDD)]] - [[Qwen]] - [[Qwen 3.8]] - [[Qwen Image 2.0]] - [[Qwen Image 3.0]] - [[Qwen3.6-27B]] - [[Qwen3.6-35B-A3B]] - [[Qwen3.8-27B]] - [[Qwen3.8-Flash-Next]] - [[RAG Pipelines]] - [[Ramp Router]] - [[Receptionist AI Design Pattern]] - [[Reinforcement Learning for Calibrated Decisions (RLCD)]] - [[Reinforcement Learning From Human Feedback (RLHF)]] - [[Reinforcement Learning with Verifiable Rewards (RLVR)]] - [[Replicate.com]] - [[Requesty]] - [[Reranking]] - [[Retrieval-Augmented Generation (RAG)]] - [[Reverse Prompter plugin for Obsidian]] - [[Reward Hacking]] - [[Roo Code]] - [[Screenshot Driven Development (SDD)]] - [[Self-Consistency]] - [[SemIf]] - [[Shadow Evaluation]] - [[ShieldGemma 2]] - [[Small Language Models (SLMs)]] - [[Smart Composer Plugin for Obsidian]] - [[Smart Connections plugin for Obsidian]] - [[Smart Second Brain plugin for Obsidian]] - [[Smart Templates plugin for Obsidian]] - [[Software 3.0]] - [[Sora]] - [[Sparse AI Models]] - [[Speculative Fan-Out]] - [[Stability AI]] - [[Stable Diffusion]] - [[SWE-Bench]] - [[System One Models]] - [[Text extractor plugin for Obsidian]] - [[Text Generator plugin for Obsidian]] - [[The limit of intelligence is contact with reality]] - [[Token Budget]] - [[Tone Matching]] - [[Treat AI context as a budget to manage]] - [[Types of Context for AI Agents]] - [[Typicality Bias]] - [[Veo 3]] - [[Vercel Open Agents]] - [[Vercel v0]] - [[Vertex AI]] - [[Video Knowledge Extraction prompt]] - [[Visual Studio Code (VSCode)]] - [[Whisk]] - [[Windsurf]] - [[Zero-Shot Classification]] - [[Zhipu AI (Z.ai)]]  ## Quotes   - [[Prompting is like playing LEGO]] - [[The more you try to outsource thinking, The less thinking you actually do]] - [[You can outsource your thinking, but you can't outsource your understanding]]