Named things
A developer trains models; a product runs on one or more models; a tool is what a builder assembles products from. 223 entries are listed as of 09/19/2026, each with the techniques it uses.
Names listed 09/19/2026. 223 of 223 registry entries have been checked against the maker's own page; the registry marks the rest as unchecked. An unchecked entry cannot be cited on a measured page until it has a primary source, a checked date, and is marked verified, and any entry unchecked for 90 days is flagged when the site is built. Where a technique has more than one maker, a page names at least two; the site ranks nothing and links no affiliate.
What this registry is, and is not
A technique registry, not a product comparison. If you are deciding what to pay for, these are the four things to ask the seller.
The registry lists models, developer tools and AI-first products, and records which technique each one demonstrates, with a primary source and the date it was checked. That is the whole of it. It is not a buyer’s guide: it carries no prices, no seat costs, no setup effort and no measure of quality, it ranks nothing, and it does not cover the ordinary business software a person already runs that has since grown an AI feature. Use it to work out what a product is doing underneath, not to choose which one to pay for.
If you want to buy rather than build, four things are worth asking about any product in any category. This site cannot answer them for you, because every answer depends on the plan and is the seller’s to state, but a seller who cannot answer them has told you something.
- Which level it operates at. Does it answer one request at a time, run a fixed workflow somebody wrote down, or decide its own next step? That is the question this site’s eight levels ask, and the answer sets both what the product can do without you and how many ways it can go wrong.
- Who else holds the text. The model’s maker is rarely the only company that receives what you type: the product in front of it keeps its own copy under its own terms. Who else holds the text says whose pages to read, in what order, and what the intermediaries usually do not say.
- Whether a person approves before anything is sent or changed. Ask whether an approval step exists, whether it is on by default, and what it actually covers: a message sent, a record changed, a payment made.
- Whether you can get the work back out. The documents, the records and the history, exported in a format another program can read, and what happens to them when you stop paying.
Products
67 entries from 34 makers. Things you can use. A product runs on one or more models.
- AiderAider · open-source coding agentCoding agents
- Antigravity CLIGoogle · coding agent · formerly Gemini CLICoding agents, Skills
- ChatGPTOpenAI · chat appChat, Images, audio and video
- ChatGPT data analysisOpenAI · code execution in a chat appCode execution
- ChatGPT deep researchOpenAI · research agentAgentic RAG and deep research
- ChatGPT memoryOpenAI · memory in a chat appMemory
- ChatGPT ProjectsOpenAI · files and instructions in a chat appContext engineering, Retrieval-augmented generation (RAG)
- ChatGPT voiceOpenAI · voice assistantVoice agents
- ChatGPT WorkOpenAI · always-on agentAlways-on assistants, Computer and browser use
- ClaudeAnthropic · chat appChat, Images, audio and video, Always-on assistants
- Claude CodeAnthropic · coding agentCoding agents, Skills, The agent harness
- Claude Code subagentsAnthropic · multi-agent feature of a coding agentLead agent and workers, Parallel calls
- Claude connectorsAnthropic · tool connections in a chat appModel Context Protocol, Function calling
- Claude CoworkAnthropic · always-on agentAlways-on assistants
- Claude memoryAnthropic · memory in a chat appMemory
- Claude ProjectsAnthropic · files and instructions in a chat appContext engineering, Retrieval-augmented generation (RAG)
- Claude ResearchAnthropic · research agentAgentic RAG and deep research, Lead agent and workers
- ClineCline · open-source coding agentCoding agents, Human approval, Deciding what to hand over
- CodexOpenAI · coding agentCoding agents, Computer and browser use, The agent harness
- CursorAnysphere · coding agent in an editorCoding agents, Chat
- Custom GPTs with actionsOpenAI · tool calling in a chat appFunction calling
- DeepSeekDeepSeek · chat appChat
- DevinCognition · coding agentCoding agents, Long-running tasks
- Devin DesktopCognition · coding agent in an editor · formerly WindsurfCoding agents, Lead agent and workers
- DifyLangGenius · visual workflow builderWorkflow graphs, Retrieval-augmented generation (RAG)
- ElevenLabsElevenLabs · voice generationImages, audio and video, Voice agents
- FigureFigure AI · humanoid robotRobots and machines
- FlowiseFlowise · visual workflow builderWorkflow graphs
- GeminiGoogle · chat appChat, Images, audio and video
- Gemini connected appsGoogle · tool connections in a chat app · formerly Gemini ExtensionsFunction calling
- Gemini Deep ResearchGoogle · research agentAgentic RAG and deep research
- Gemini LiveGoogle · voice assistantVoice agents
- Gemini NotebookGoogle · research notebook · formerly NotebookLMRetrieval-augmented generation (RAG), Code execution
- Gemini SparkGoogle · always-on agentAlways-on assistants
- GitHub CopilotGitHub · coding agent in an editorCoding agents, Chat
- GleanGlean · workplace searchRetrieval-augmented generation (RAG)
- Glean Enterprise GraphGlean · knowledge graph inside a workplace search productKnowledge graphs and GraphRAG
- GrokSpaceXAI · chat appChat
- Grok BotSpaceXAI · always-on agentAlways-on assistants
- Grok BuildSpaceXAI · coding agentCoding agents, Lead agent and workers
- Grok DeepSearchSpaceXAI · research agentAgentic RAG and deep research
- Grok HeavySpaceXAI · several agents answering one questionReview and debate, Parallel calls
- GumloopGumloop · visual workflow builderWorkflow graphs, Model Context Protocol
- Hermes AgentNous Research · always-on agent, self-hostedAlways-on assistants
- JulesGoogle · coding agentCoding agents, Human approval, Reviewing work you did not do
- LangflowIBM · visual workflow builderWorkflow graphs, Model Context Protocol
- LinkStripe · digital wallet for checkoutAlways-on assistants
- MakeCelonis · automation servicePrompt chaining
- ManusManus · general-purpose agentLong-running tasks
- Meta AIMeta · chat appChat
- Microsoft 365 CopilotMicrosoft · workplace assistantRetrieval-augmented generation (RAG)
- Microsoft 365 Copilot semantic indexMicrosoft · vector index behind a workplace assistantEmbeddings and search
- Microsoft CopilotMicrosoft · chat appChat
- MidjourneyMidjourney · image generationImages, audio and video
- Mistral VibeMistral AI · ai agent for work and coding · formerly Le ChatChat, Function calling, Code execution
- MuseMeta · always-on agentAlways-on assistants, Review and debate
- n8nn8n · automation service, self-hostablePrompt chaining, Workflow graphs, Code execution
- NEO1X · home robotRobots and machines
- Notion AINotion · workspace assistantRetrieval-augmented generation (RAG)
- OpenClawOpenClaw Foundation · always-on agent, self-hostedAlways-on assistants, Long-running tasks
- OptimusTesla · humanoid robotRobots and machines
- PerplexityPerplexity · answer engineRetrieval-augmented generation (RAG)
- Perplexity Deep ResearchPerplexity · research agentAgentic RAG and deep research
- Power AutomateMicrosoft · automation servicePrompt chaining, Human approval
- Replit AgentReplit · coding agentCoding agents, Parallel calls
- ZapierZapier · automation servicePrompt chaining, Routing
- ZedZed Industries · ai code editorCoding agents
Tools
113 entries from 66 makers. Things you build with: frameworks, runtimes, protocols and libraries.
- Agent Development KitGoogle · agent frameworkSingle agent, Agent graphs, The agent harness
- Agent SkillsAnthropic · format for reusable agent instructionsSkills
- Agent2Agent (A2A) Protocolopen standard · protocol between agentsAgent graphs
- AGENTS.mdopen convention · instructions file for coding agentsSkills
- AI Guardrails (Lakera Guard)Check Point · prompt-injection filter · formerly Lakera GuardSafety, privacy and governance, Guardrails
- AI SDKVercel · TypeScript AI and agent SDKChat, Single agent, Function calling, Structured output
- Apache Airflowopen source · workflow enginePrompt chaining
- AutoGenMicrosoft · multi-agent frameworkLead agent and workers, Review and debate
- Axolotlopen source · fine-tuning libraryChanging the model, Fine-tuning and adapters
- Batch APIOpenAI · asynchronous batch processing apiParallel calls
- Batch APIGoogle · asynchronous batch processing api · formerly Batch ModeParallel calls
- BraintrustBraintrust · eval platformEvals, Evaluation frameworks
- Browser UseBrowser Use · browser agent frameworkComputer and browser use, Single agent
- BrowserbaseBrowserbase · hosted browsers for agentsComputer and browser use
- ChromaChroma · vector databaseEmbeddings and search
- Claude Agent SDKAnthropic · agent frameworkSingle agent, Lead agent and workers, The agent harness
- Claude APIAnthropic · model APIChat, Structured output, Function calling
- Claude computer useAnthropic · computer-use APIComputer and browser use
- Cloudflare AI GatewayCloudflare · hosted gateway for model callsAI gateways, Cost optimization
- Cohere RerankCohere · rerankerRetrieval-augmented generation (RAG), Embeddings and search
- ComposioComposio · prebuilt tool connectionsFunction calling, Model Context Protocol
- CrewAICrewAI · multi-agent frameworkLead agent and workers, Agent graphs, The agent harness
- Deep AgentsLangChain · agent harnessThe agent harness, Single agent, Context engineering, Memory, Skills, Lead agent and workers, Human approval
- DeepEvalConfident AI · eval frameworkEvals, Evaluation frameworks
- distilabelArgilla · synthetic dataChanging the model, Synthetic data
- DSPyStanford NLP · prompt programs and optimizersWrite and check, Changing the model, Prompt optimization
- E2BE2B · code sandboxCode execution
- ElasticsearchElastic · keyword searchWhen not to use a model
- FAISSMeta · vector search libraryEmbeddings and search
- garakNVIDIA · LLM vulnerability scannerRed teaming, Safety, privacy and governance
- Gemini APIGoogle · model APIChat, Structured output, Function calling
- Gemini computer useGoogle · computer-use APIComputer and browser use
- Gemini Live APIGoogle · voice APIVoice agents
- GraphRAGMicrosoft · knowledge-graph retrievalKnowledge graphs and GraphRAG
- Guardrails AIGuardrails AI · guardrails frameworkSafety, privacy and governance, Guardrails
- Haystackdeepset · retrieval frameworkRetrieval-augmented generation (RAG), Prompt chaining
- HeliconeHelicone · tracing and cost trackingOperations, Observability, Cost optimization
- InngestInngest · durable workflow enginePrompt chaining, Long-running tasks
- InspectUK AI Security Institute · eval frameworkEvals
- Instructoropen source · structured output libraryStructured output
- jina-reranker-v3.5Jina AI · rerankerRetrieval-augmented generation (RAG), Embeddings and search
- LangChainLangChain · application frameworkRetrieval-augmented generation (RAG), Prompt chaining, Parallel calls
- LangfuseClickHouse · tracing and cost trackingOperations, Observability, Cost optimization
- LangGraphLangChain · graph frameworkWorkflow graphs, Agent graphs, Human approval, The agent harness
- LangSmithLangChain · eval and tracing platformEvals, Operations, Observability, Evaluation frameworks
- LangSmith DeploymentLangChain · hosting for long-running agents · formerly LangGraph PlatformLong-running tasks
- LazyGraphRAGMicrosoft · knowledge-graph retrievalKnowledge graphs and GraphRAG
- LettaLetta · agents with long-term memory · formerly MemGPTMemory, Long-running tasks
- LightRAGopen source · knowledge-graph retrievalKnowledge graphs and GraphRAG
- LiteLLMBerriAI · one API for many modelsChat, Operations, AI gateways
- LiveKit AgentsLiveKit · voice agent frameworkVoice agents
- Llama Guard 4Meta · safety classifier · formerly Llama GuardSafety, privacy and governance, Guardrails
- llama.cppopen source · runs models locallyChat, Operations, Running models locally
- LlamaIndexLlamaIndex · retrieval frameworkRetrieval-augmented generation (RAG), Agentic RAG and deep research
- LM StudioElement Labs · runs models locallyChat, Operations, Running models locally
- lm-evaluation-harnessEleutherAI · benchmark runnerEvals
- Mem0Mem0 · memory layerMemory
- Message Batches APIAnthropic · asynchronous batch processing apiParallel calls
- MetaGPTopen source · multi-agent frameworkOrganizations of agents
- Microsoft Agent FrameworkMicrosoft · multi-agent frameworkAgent graphs, The agent harness
- MilvusZilliz · vector databaseEmbeddings and search
- MLXApple · training and inference on Apple hardwareChanging the model, Operations, Running models locally, Fine-tuning and adapters
- ModalModal · code sandbox and computeCode execution
- Model Context Protocolopen standard · protocol for tools and dataModel Context Protocol
- NeMo GuardrailsNVIDIA · guardrails frameworkSafety, privacy and governance, Guardrails
- Neo4jNeo4j · graph databaseKnowledge graphs and GraphRAG
- OllamaOllama · runs models locallyChat, Operations, Running models locally
- OpenAI Agents SDKOpenAI · agent frameworkSingle agent, Agent graphs, The agent harness
- OpenAI APIOpenAI · model APIChat, Structured output, Function calling
- OpenAI computer useOpenAI · computer-use APIComputer and browser use
- OpenAI EvalsOpenAI · eval frameworkEvals
- OpenAI fine-tuningOpenAI · hosted fine-tuningChanging the model, Fine-tuning and adapters
- OpenAI Realtime APIOpenAI · voice APIVoice agents
- OpenRouterOpenRouter · one API for many modelsChat, Operations, AI gateways
- OpenSearchopen source · keyword searchWhen not to use a model
- OpenTelemetryopen standard · tracing standardOperations, Observability
- Outlinesdottxt · structured output libraryStructured output
- PEFTHugging Face · LoRA and other adaptersChanging the model, Fine-tuning and adapters
- pgvectoropen source · vector search in PostgresEmbeddings and search
- PhoenixArize AI · AI observability and evaluationObservability, Evals, Evaluation frameworks, Operations
- PineconePinecone · vector databaseEmbeddings and search
- PipecatDaily · voice agent frameworkVoice agents
- PlaywrightMicrosoft · browser automationComputer and browser use
- PrefectPrefect · workflow enginePrompt chaining
- Prompt Design StrategiesGoogle · prompt engineering guidePrompt engineering
- Prompt EngineeringOpenAI · prompt engineering guidePrompt engineering
- Prompt Engineering OverviewAnthropic · prompt engineering guidePrompt engineering
- Prompt GeneratorAnthropic · prompt generation toolPrompt engineering
- Prompt ImproverAnthropic · prompt optimization toolPrompt engineering
- promptfoopromptfoo · eval runnerEvals, Evaluation frameworks
- PydanticPydantic · schema validationStructured output
- Pydantic AIPydantic · agent frameworkSingle agent, The agent harness
- PyRITMicrosoft · red-teaming frameworkRed teaming
- QdrantQdrant · vector databaseEmbeddings and search
- Ragasopen source · evals for retrievalEvals, Evaluation frameworks
- Regular expressionsevery language · pattern matchingWhen not to use a model
- scikit-learnopen source · classical machine learningWhen not to use a model
- Semantic RouterAurelio Labs · library that routes a request by meaningRouting
- SGLangSGLang community · model serving runtimeRunning models locally, Operations
- smolagentsHugging Face · agent frameworkSingle agent
- spaCyExplosion · rule-based and statistical text processingWhen not to use a model
- SQLevery database · structured queriesWhen not to use a model
- Strands AgentsStrands Agents · agent harness SDKThe agent harness, Single agent, Function calling, Model Context Protocol, Agent graphs
- TemporalTemporal · durable workflow enginePrompt chaining, Long-running tasks
- Together AI fine-tuningTogether AI · hosted fine-tuningChanging the model, Fine-tuning and adapters
- TransformersHugging Face · model libraryChat, Changing the model, Fine-tuning and adapters
- TRLHugging Face · fine-tuning libraryChanging the model, Fine-tuning and adapters
- UnslothUnsloth · fine-tuning libraryChanging the model, Fine-tuning and adapters
- VapiVapi · voice agent platformVoice agents
- vLLMopen source · model serverOperations, Running models locally
- WeaviateWeaviate · vector databaseEmbeddings and search
- XGBoostopen source · classical machine learningWhen not to use a model
- ZepZep · memory layerMemory
Models
43 entries from 22 makers. Things a developer trains. A product runs on one or more of these.
- BGEBAAI · open embedding modelEmbeddings and search
- Claude Haiku 4.5Anthropic · small modelChat
- Claude Opus 5Anthropic · frontier modelChat, Reasoning at answer time
- Claude Sonnet 5Anthropic · mid-size modelChat
- Cohere Embed v4Cohere · embedding model · formerly Cohere Embed, genericEmbeddings and search
- Command A+Cohere · enterprise model · formerly Command, deprecated September 15, 2025Chat
- DeepSeek V4DeepSeek · open-weight modelChat, Reasoning at answer time
- DeepSeek-V4.1-FlashDeepSeek · open-weight modelChat, Reasoning at answer time
- FLUX 3Black Forest Labs · image and video generation model · formerly FLUX, superseded July 23, 2026Images, audio and video
- Gemini 3.1 ProGoogle · frontier modelChat, Reasoning at answer time
- Gemini 3.5 Flash-LiteGoogle · small modelChat
- Gemini 3.8 FlashGoogle · mid-size modelChat
- Gemini Robotics 2Google · robotics model · formerly Gemini Robotics, supersededRobots and machines
- Gemma 4Google · open-weight model · formerly GemmaChat
- Gen-4.5Runway · text-to-video modelImages, audio and video
- GLM-5.3Z.ai · open-weight modelChat
- GPT-5.6 LunaOpenAI · small modelChat
- GPT-5.6 SolOpenAI · frontier modelChat
- GPT-5.6 TerraOpenAI · mid-size modelChat
- GPT-6 AstraOpenAI · frontier modelChat, Reasoning at answer time
- gpt-ossOpenAI · open-weight modelChat
- GR00TNVIDIA · robotics modelRobots and machines
- Grok 4.6SpaceXAI · frontier modelChat, Reasoning at answer time
- Imagen 4Google · image model · formerly Imagen, genericImages, audio and video
- JevTypeSafe AI · system one decision modelStructured output, Routing, Guardrails, Parallel calls
- Kimi K3Moonshot AI · open-weight modelChat
- Llama 4Meta · open-weight model · formerly LlamaChat
- Mistral Large 3Mistral AI · open-weight modelChat
- Mistral Small 4Mistral AI · open-weight modelChat
- Muse Spark 1.3Meta · frontier model · formerly Muse Spark, 2026-04Chat
- Nemotron 3NVIDIA · open-weight model · formerly Nemotron, superseded December 15, 2025Chat
- Nomic Embed Text v2Nomic · open embedding model · formerly Nomic EmbedEmbeddings and search
- Phi-4-miniMicrosoft · small open-weight model · formerly PhiChat
- Qwen3.8Alibaba · open-weight modelChat
- Sora 2OpenAI · video model · formerly Sora, superseded September 30, 2025Images, audio and video
- Stable Diffusion 3.5Stability AI · text-to-image modelImages, audio and video
- SunoSuno · text-to-music modelImages, audio and video
- text-embedding-3OpenAI · embedding modelEmbeddings and search
- Veo 3.1Google · video model · formerly Veo, supersededImages, audio and video
- Voyage Embed 4Voyage AI · embedding model · formerly Voyage embeddingsEmbeddings and search
- WhisperOpenAI · speech-to-text modelImages, audio and video
- π0Physical Intelligence · robotics modelRobots and machines
- π0.7Physical Intelligence · robotics modelRobots and machines