llms.txt
Compare with other sites' llms.txt files →# Product Builder
> Hand-picked podcasts, books, tools, models, MCP servers and guides for people who build products with AI. Model and MCP entries list the date they were checked and link to the provider's own documentation.
## Learn guides
Practical guides for building products with AI agents, models and no-code tools.
- [Always-On AI Agents Compared: Duet, Dots, Grok Bot, Muse & Spark](https://www.productbuilder.net/learn/always-on-agents): Claude Duet vs OpenAI Dots vs Grok Bot vs Meta Muse vs Gemini Spark: what always-on AI agents do, who can get them, and the approval controls to check.
- [Decision Models & APIs: Microsoft, Perplexity, Jev & OpenAI](https://www.productbuilder.net/learn/decision-models): Compare Microsoft-Decision-1, pplx-decider, Jev, and OpenAI Decisions, then design safe thresholds, abstention, and human review.
- [Claude 5.5 Models: Haiku vs Sonnet vs Opus](https://www.productbuilder.net/learn/claude-5-5-models): Choose Claude Haiku 5.5, Sonnet 5.5, or Opus 5.5 by workload, risk, and real task cost—not benchmark rank alone.
- [What Is a Product Builder?](https://www.productbuilder.net/learn/what-is-a-product-builder): The AI-native role explained: someone who decides what to build and builds it, pairing product judgment with AI coding agents and no-code tools. How it differs from a product manager, and how to become one.
- [What Is a Product Manager?](https://www.productbuilder.net/learn/what-is-a-product-manager): A product manager decides what a team builds and why. What PMs actually do, PM vs. project manager, the skills that matter, and how AI is changing the job.
- [What Is a Product Engineer?](https://www.productbuilder.net/learn/what-is-a-product-engineer): A product engineer is a software engineer who owns the outcome, not just the code. What they do, how the role differs, and how AI widens it.
- [What Is an AI Engineer?](https://www.productbuilder.net/learn/what-is-an-ai-engineer): An AI engineer builds products on top of AI models: prompts, RAG, agents, and evals. What they do, AI engineer vs. ML engineer, and how to become one.
- [What Is an AI Product Manager?](https://www.productbuilder.net/learn/what-is-an-ai-product-manager): An AI product manager decides what AI features to build and what “good” means for them. What AI PMs do, AI PM vs. PM, and the skills that matter.
- [What Is a Design Engineer?](https://www.productbuilder.net/learn/what-is-a-design-engineer): A design engineer designs and builds the interface with taste and code. What they do, design engineer vs. UX engineer vs. front-end, and how to start.
- [What Is a Forward Deployed Engineer?](https://www.productbuilder.net/learn/what-is-a-forward-deployed-engineer): A forward deployed engineer embeds with customers to make a product work for them. What FDEs do, FDE vs. solutions engineer, and why AI labs hire them.
- [What Is a Technical Program Manager?](https://www.productbuilder.net/learn/what-is-a-technical-program-manager): What a technical program manager (TPM) does, TPM vs product manager vs engineering manager, the system design interview, and how AI changes the job.
- [What Is a Founding Engineer?](https://www.productbuilder.net/learn/what-is-a-founding-engineer): A founding engineer is one of a startup's first engineers. What the job is, founding engineer vs. CTO vs. early engineer, equity, and what to ask.
- [What Is a Growth Engineer?](https://www.productbuilder.net/learn/what-is-a-growth-engineer): What is a growth engineer? An engineer who ships experiments on sign-up, activation, and revenue. Growth engineer vs. product engineer vs. growth marketer.
- [Product Manager vs Product Engineer vs Product Builder](https://www.productbuilder.net/learn/product-manager-vs-product-engineer-vs-product-builder): Product manager vs product engineer vs product builder: who owns what, who writes code, how they share one team, and which role fits you.
- [What Is Vibe Coding?](https://www.productbuilder.net/learn/vibe-coding): Building software by describing what you want in plain English and letting AI write the code. Where Karpathy's term came from, what it's great for, when it bites, and how to graduate to spec-driven development.
- [Spec-Driven Development](https://www.productbuilder.net/learn/spec-driven-development): Write the spec before the prompt: how structured specifications make AI coding agents like Claude Code, GitHub Spec Kit, and Kiro produce production-ready code. What it is, when to use it, and how to start.
- [Context Engineering](https://www.productbuilder.net/learn/context-engineering): The meta-skill of building with AI: deciding what the model can see — instructions, tools, memory, retrieved docs, history — not just how you phrase the prompt. Covers context rot, retrieval vs stuffing, and context vs prompt engineering.
- [Harness Engineering](https://www.productbuilder.net/learn/harness-engineering): The system around the model that makes an agent reliable: the loop, verification, state, tools, human-in-the-loop, and lifecycle. The model is the CPU; the harness is the OS. The prompt → context → harness arc.
- [AI Evaluations](https://www.productbuilder.net/learn/ai-evaluations): Test and validate AI systems before shipping. The #1 skill for 2025 according to CPOs of OpenAI and Anthropic.
- [Model Context Protocol (MCP)](https://www.productbuilder.net/learn/model-context-protocol): Standardize how AI agents connect to tools and data sources. The future of agent interoperability.
- [Prompt Engineering](https://www.productbuilder.net/learn/prompt-engineering): Craft effective prompts that get better results from AI models
- [Chain-of-Thought](https://www.productbuilder.net/learn/chain-of-thought): Improve AI reasoning by asking models to think step-by-step
- [Function Calling](https://www.productbuilder.net/learn/function-calling): Enable AI agents to use external tools and APIs
- [Agent Memory](https://www.productbuilder.net/learn/agent-memory): Give your AI agents the ability to remember context across conversations
- [Implementing Episodic Memory](https://www.productbuilder.net/learn/implementing-episodic-memory): Production-ready guide with code examples for agent memory
- [Vector Databases](https://www.productbuilder.net/learn/vector-databases): Storage for agent memory and semantic search
- [RAG Systems](https://www.productbuilder.net/learn/rag-systems): Give AI models access to external knowledge through RAG
- [Agentic Workflows](https://www.productbuilder.net/learn/agentic-workflows): Design patterns for building autonomous AI agents
- [Context Optimization](https://www.productbuilder.net/learn/context-optimization): Optimize context windows for better performance
- [LLM Security](https://www.productbuilder.net/learn/llm-security): Security risks and best practices for AI applications
- [Prompt Caching](https://www.productbuilder.net/learn/prompt-caching): Reduce costs by 90% by caching repeated context
- [Company Strategies](https://www.productbuilder.net/learn/company-strategies): How leading companies build AI products
- [Open vs Proprietary](https://www.productbuilder.net/learn/open-vs-proprietary): Choosing between open-source and proprietary models
- [RAG vs Long Context](https://www.productbuilder.net/learn/rag-vs-long-context): When to use RAG vs long context windows
- [Claude Code: Complete Guide](https://www.productbuilder.net/learn/claude-code): Master Claude Code from beginner to advanced. Skills, CLAUDE.md, MCP servers, hooks, and autonomous workflows.
- [OpenAI Codex](https://www.productbuilder.net/learn/codex): A plain-English guide to OpenAI Codex for product builders: what it is, what you can build with it, whether it's for you, and how to start in five minutes.
- [Claude Code Tips & Workflows](https://www.productbuilder.net/learn/claude-code-tips): 40+ expert tips for prompting, planning, parallelism, debugging, and daily habits from Claude Code power users.
- [CLAUDE.md & Memory Management](https://www.productbuilder.net/learn/claude-code-memory): Write effective CLAUDE.md files, understand monorepo memory loading, manage persistent memory, rules, and the @ import syntax.
- [Claude Code Commands, Skills & Agents](https://www.productbuilder.net/learn/claude-code-skills): Master the Claude Code extension system: slash commands, reusable skills, custom subagents, and orchestration workflows.
- [Claude Code Settings & Configuration](https://www.productbuilder.net/learn/claude-code-settings): Complete reference for Claude Code settings hierarchy, permissions, hooks, MCP servers, sandboxing, and 140+ environment variables.
## AI models
Model pages with context window, output limit, pricing and the date each was checked against the provider.
- [Gemini 4 Argon](https://www.productbuilder.net/models/gemini-4-argon): Google | coding | 1M tokens context | $2.00/M tokens input / $10.00/M tokens output (per million tokens)
- [GPT-6.1 Sol](https://www.productbuilder.net/models/gpt-6-1-sol): OpenAI | coding | 1.05M tokens context | $2.00/M tokens input / $10.00/M tokens output (per million tokens)
- [GPT-6.1 Terra](https://www.productbuilder.net/models/gpt-6-1-terra): OpenAI | language | 1.05M tokens context | $1.00/M tokens input / $5.00/M tokens output (per million tokens)
- [Mistral Large 4](https://www.productbuilder.net/models/mistral-large-4): Mistral AI | language | 1M tokens context | $1.36/M tokens input / $4.18/M tokens output (per million tokens)
- [Gemini Nano Banana 2.1](https://www.productbuilder.net/models/gemini-nano-banana-2-1): Google | image generation | 131K input context | $1.50/M tokens input / $30 image / $7.50 text/M tokens output (per million tokens)
- [EmbeddingGemma 2](https://www.productbuilder.net/models/embeddinggemma-2): Google | embeddings | 8,192 tokens context | Self-hosted input / N/A output (variable by output settings)
- [pplx-decider-v1.1-27b](https://www.productbuilder.net/models/pplx-decider-v1-1-27b): pplx-decider-v1.1-27b is a gated multimodal decision model derived from Qwen3.8-27B. Review its 8K limit, hardware cost, and safeguards.
- [Claude Sonnet 5.5](https://www.productbuilder.net/models/claude-sonnet-5-5): Anthropic | coding | 1M tokens context | $2.00/M tokens input / $10.00/M tokens output (per million tokens)
- [DeepSeek V4.1 Flash](https://www.productbuilder.net/models/deepseek-v4-1-flash): DeepSeek | language | 1M tokens context | See provider input / See provider output (variable by output settings)
- [Qwen3.8 Omni Flash](https://www.productbuilder.net/models/qwen3-8-omni-flash): Alibaba | audio | 1M tokens context | See provider input / See provider output (variable by output settings)
- [Qwen3.8 Omni Flash Realtime](https://www.productbuilder.net/models/qwen3-8-omni-flash-realtime): Alibaba | realtime voice | 196K input tokens context | $0.23/M tokens input / $0.70/M tokens output (per million tokens)
- [Microsoft-Decision-1](https://www.productbuilder.net/models/microsoft-decision-1): Microsoft | decision | 32K tokens context | $0.042/M tokens input / $0/M tokens output (per million tokens)
- [Jev](https://www.productbuilder.net/models/jev): Jev turns text into typed choices, scores, and yes/no probabilities for software decisions. See its early-access limits, pricing, and a realistic pilot.
- [Gemini 3.8 Live](https://www.productbuilder.net/models/gemini-3-8-live): Google | realtime voice | 131K tokens context | $0.75/M tokens input / $4.50/M tokens output (per million tokens)
- [Gemini 3.8 Live Extended Thinking](https://www.productbuilder.net/models/gemini-3-8-live-extended-thinking): Google | realtime voice | 131K tokens context | $0.75/M tokens input / $4.50/M tokens output (per million tokens)
- [Claude Opus 5.5](https://www.productbuilder.net/models/claude-opus-5-5): Anthropic | coding | 1M tokens context | $4.00/M tokens input / $20.00/M tokens output (per million tokens)
- [GPT-6 Sol](https://www.productbuilder.net/models/gpt-6-sol): OpenAI | coding | 1.05M tokens context | $2.00/M tokens input / $10.00/M tokens output (per million tokens)
- [Grok 4.7](https://www.productbuilder.net/models/grok-4-7): SpaceXAI | coding | 500K tokens context | $2.00/M tokens input / $6.00/M tokens output (per million tokens)
- [MiMo V2.6 Pro](https://www.productbuilder.net/models/mimo-v2-6-pro): Xiaomi | language | 1M tokens context | $0.435/M tokens input / $0.87/M tokens output (per million tokens)
- [MiMo V2.6 Flash](https://www.productbuilder.net/models/mimo-v2-6-flash): Xiaomi | language | 1M tokens context | $0.14/M tokens input / $0.28/M tokens output (per million tokens)
- [Claude Haiku 5.5](https://www.productbuilder.net/models/claude-haiku-5-5): Anthropic | language | 1M tokens context | $0.10/M tokens input / $0.50/M tokens output (per million tokens)
- [GPT-6 Luna](https://www.productbuilder.net/models/gpt-6-luna): OpenAI | language | 1.05M tokens context | $0.10/M tokens input / $0.50/M tokens output (per million tokens)
- [GPT-Realtime-2.1](https://www.productbuilder.net/models/gpt-realtime-2-1): OpenAI | realtime voice | 128K tokens context | $4.00/M tokens input / $24.00/M tokens output (per million tokens)
- [GPT-Image-2](https://www.productbuilder.net/models/gpt-image-2): OpenAI | image generation | Image prompt context | Token-based input / Varies output (variable by output settings)
- [GPT-Audio-1.5](https://www.productbuilder.net/models/gpt-audio-1-5): OpenAI | audio | 128K tokens context | $2.50/M tokens input / $10.00/M tokens output (per million tokens)
- [Muse Video](https://www.productbuilder.net/models/muse-video): Meta | video generation | Not published context | Not published input / Not published output (provider pricing not published)
- [GPT-6 Astra](https://www.productbuilder.net/models/gpt-6-astra): OpenAI | language | 1.05M tokens context | $10.00/M tokens input / $50.00/M tokens output (per million tokens)
- [Lyria 3.5](https://www.productbuilder.net/models/lyria-3-5-pro): Google | music generation | 131K tokens context | $0.08/request input / Included output (per request)
- [GPT-5.6 Cyber](https://www.productbuilder.net/models/gpt-5-6-cyber): OpenAI | cybersecurity | 400K tokens context | $12.50/M tokens input / $75.00/M tokens output (per million tokens)
- [Qwen 3.8 Max](https://www.productbuilder.net/models/qwen-3-8-max): Alibaba | language | 1M tokens context | $2.00/M tokens input / $6.00/M tokens output (per million tokens)
- [Qwen3.8 2.4T A95B](https://www.productbuilder.net/models/qwen3-8-2-4t-a95b): Alibaba | language | 1M tokens context | $2.00/M tokens input / $6.00/M tokens output (per million tokens)
- [Gemini 3.5 Transcribe Live](https://www.productbuilder.net/models/gemini-3-5-transcribe-live): Google | speech to text | Streaming audio context | $3.50/M tokens input / $21.00/M tokens output (per million tokens)
- [MAI-Image-2.6](https://www.productbuilder.net/models/mai-image-2-6): Microsoft | image generation | Not published context | Not published input / Not published output (provider pricing not published)
- [Claude Fable 5.1](https://www.productbuilder.net/models/claude-fable-5-1): Anthropic | language | 1M tokens context | $10.00/M tokens input / $50.00/M tokens output (per million tokens)
- [Claude Mythos 5.1](https://www.productbuilder.net/models/claude-mythos-5-1): Anthropic | cybersecurity | 1M tokens context | $10.00/M tokens input / $50.00/M tokens output (per million tokens)
- [Grok 4.6](https://www.productbuilder.net/models/grok-4-6): xAI | language | 500K tokens context | $2.00/M tokens input / $6.00/M tokens output (per million tokens)
- [Claude Fable 5](https://www.productbuilder.net/models/claude-fable-5): Anthropic | language | 1M tokens context | $10.00/M tokens input / $50.00/M tokens output (per million tokens)
- [Claude Opus 5](https://www.productbuilder.net/models/claude-opus-5): Anthropic | language | 1M tokens context | $5.00/M tokens input / $25.00/M tokens output (per million tokens)
- [Claude Opus 4.8](https://www.productbuilder.net/models/claude-opus-4-8): Anthropic | language | 1M tokens context | $5.00/M tokens input / $25.00/M tokens output (per million tokens)
- [Claude Opus 4.7](https://www.productbuilder.net/models/claude-opus-4-7): Anthropic | language | 1M tokens context | $5.00/M tokens input / $25.00/M tokens output (per million tokens)
- [GPT-5.6 Sol](https://www.productbuilder.net/models/gpt-5-6-sol): OpenAI | language | 1.05M tokens context | $4.00/M tokens input / $20.00/M tokens output (per million tokens)
- [GPT-5.5](https://www.productbuilder.net/models/gpt-5-5): OpenAI | language | 1.05M tokens context | $5.00/M tokens input / $30.00/M tokens output (per million tokens)
- [GPT-5.5 Pro](https://www.productbuilder.net/models/gpt-5-5-pro): OpenAI | language | 1.05M tokens context | $30.00/M tokens input / $180.00/M tokens output (per million tokens)
- [GPT-5.4](https://www.productbuilder.net/models/gpt-5-4-thinking): OpenAI | language | 1.05M tokens context | $2.50/M tokens input / $15.00/M tokens output (per million tokens)
- [GPT-5.4 Pro](https://www.productbuilder.net/models/gpt-5-4-pro): OpenAI | language | 1.05M tokens context | $30.00/M tokens input / $180.00/M tokens output (per million tokens)
- [Claude Opus 4.6](https://www.productbuilder.net/models/claude-opus-4-6): Anthropic | language | 1M tokens context | $5.00/M tokens input / $25.00/M tokens output (per million tokens)
- [GPT-5.3 Codex](https://www.productbuilder.net/models/gpt-5-3-codex): OpenAI | coding | 400K tokens context | $1.75*/M tokens input / $14.00*/M tokens output (per million tokens)
- [GPT-5.3-Codex-Spark](https://www.productbuilder.net/models/gpt-5-3-codex-spark): OpenAI | coding | 128K tokens context | Credit-based* input / Credit-based* output (per million tokens)
- [Gemini 3.5 Pro](https://www.productbuilder.net/models/gemini-3-5-pro): Google | language | 2M tokens context | TBD* input / TBD* output (per million tokens)
- [Gemini Omni Flash](https://www.productbuilder.net/models/gemini-omni): Google | video generation | 1M tokens context | $1.50/M tokens input / $17.50 video*/M tokens output (per million tokens)
- [Gemini 3.1 Pro](https://www.productbuilder.net/models/gemini-3-1-pro): Google | language | 1M tokens context | $2.00*/M tokens input / $12.00*/M tokens output (per million tokens)
- [Gemini 3 Pro Preview](https://www.productbuilder.net/models/gemini-3-pro): Google | language | 1M tokens context | $2.00*/M tokens input / $12.00*/M tokens output (per million tokens)
- [Kimi K3](https://www.productbuilder.net/models/kimi-k3): Moonshot AI | language | 1M tokens context | $3.00/M tokens input / $15.00/M tokens output (per million tokens)
- [Kimi K2.6](https://www.productbuilder.net/models/kimi-k2-6): Moonshot AI | language | 256K tokens context | $0.80/M tokens input / $3.50/M tokens output (per million tokens)
- [Kimi K2 Thinking](https://www.productbuilder.net/models/kimi-k2-thinking): Moonshot AI | language | 256K tokens context | $0.80/M tokens input / $3.20/M tokens output (per million tokens)
- [Claude Opus 4.5](https://www.productbuilder.net/models/claude-opus-4-5): Anthropic | language | 200K tokens context | $5.00/M tokens input / $25.00/M tokens output (per million tokens)
- [Claude Opus 4.1](https://www.productbuilder.net/models/claude-opus-4-1): Anthropic | language | 200K tokens context | $15.00/M tokens input / $75.00/M tokens output (per million tokens)
- [Gemini 2.5 Pro](https://www.productbuilder.net/models/gemini-2-5-pro): Google | language | 1M tokens context | $1.25*/M tokens input / $10.00*/M tokens output (per million tokens)
- [GPT-5.2 Pro](https://www.productbuilder.net/models/gpt-5-2-pro): OpenAI | language | 400K tokens context | $21.00/M tokens input / $168.00/M tokens output (per million tokens)
- [GPT-5.2 Thinking](https://www.productbuilder.net/models/gpt-5-2-thinking): OpenAI | language | 400K tokens context | $1.75/M tokens input / $14.00/M tokens output (per million tokens)
- [GPT-5.2 Instant](https://www.productbuilder
