Browse entity types
AI entities, people, products, features, and tools
Explore named AI products, models, people, frameworks, features, and tools. Each entry brings together what it is, why it matters, official sources, and useful next steps.
Entity type
Models
01Image generation and editingSeedream 5.5Explore Seedream 5.5’s precise image edits, 100-round demonstration, native transparent assets, access options and API costs.02Coding and agent modelsGemini 4 ArgonGoogle's frontier model with October 13 paid API and Pro/Ultra app rollout; Fairwind cyber access remains separate.03Coding and agent modelsGPT-6.1 SolOpenAI's September 2026 coding and agent model, available in the API and paid ChatGPT Work and Codex plans.04Decision modelsJevTypeSafe’s text-only System One decision model: versioned Jev 1.14, stable/preview aliases, typed probabilities and per-request/question limits.05AI model APIsGemini 3.5 FlashGoogle's previous Gemini Flash-tier model for agentic coding, long-horizon workflows, multimodal understanding, and API use.06AI modelsGemini 3.5 ProGoogle frontier Gemini model in partner testing, with transient Arena observations but no documented stable public API release yet.07AI model APIsGemini 3.6 FlashGoogle's GA Flash workhorse for coding, multimodal work, and agentic workflows with lower output pricing than Gemini 3.5 Flash.08AI model APIsGemini 3.5 Flash-LiteGoogle's fastest and lowest-cost Gemini 3.5 model for high-throughput agents, search, extraction, and document processing.09AI securityGemini 3.5 Flash CyberGoogle DeepMind's security-specialist Gemini model for finding, validating, and patching software vulnerabilities through CodeMender.10Video generationGemini OmniGoogle Gemini model family for any-input generation and conversational video creation, starting with Gemini Omni Flash.11Weather forecastingWeatherNext 3Google weather-forecasting model with live satellite inputs, hourly forecast initialization, global product integrations, and public cloud data access.12Image generationGPT Image 2Previous-generation OpenAI image model retained for API compatibility and comparison with the Images 2.5 family.13AI coding modelsComposer 2.5Cursor coding model focused on long-running agentic software tasks inside Cursor.14AI coding modelsClaude Opus 4.7Anthropic frontier Claude model for demanding coding, agent, and complex document work.15AI coding modelsClaude Opus 4.8Anthropic Opus predecessor to Claude Opus 5 for complex reasoning, long-horizon coding, and high-autonomy workflows.16AI coding modelsClaude Opus 5Anthropic Opus model for complex agentic coding, enterprise work, deep reasoning, and long-horizon tasks.17AI coding modelsClaude Fable 5Anthropic high-capability model for coding, long-horizon agentic workflows, and production-facing reasoning with safety guardrails.18AI coding modelsClaude Mythos 5Anthropic invitation-only Mythos-tier model accessible through Project Glasswing for approved customers.19AI coding modelsClaude Fable 5.1Anthropic generally available frontier model for long-horizon coding, research, and artifact work.20AI coding modelsClaude Mythos 5.1Anthropic trusted-access sibling of Fable 5.1 for approved cybersecurity and life-science work.21AI coding modelsClaude Sonnet 5Anthropic model for production coding, agent workflows, and long-context software engineering tasks.22AI coding modelsGPT-5.5OpenAI frontier model for complex professional work, coding, research, and Codex workflows.23AI coding modelsGPT-5.6OpenAI generally available model family with Sol, Terra, and Luna tiers for frontier, balanced, and cost-sensitive workloads.24AI modelsDeepSeek V4 ProDeepSeek V4 Pro endpoint on the DeepSeek-V4-Pro-0813 backend, with a current first-party conflict over whether service continues or routes to V4.1 Flash.25AI modelsDeepSeek V4 FlashRetired DeepSeek-V4-Flash-0731 model whose legacy API ID temporarily routes to the multimodal V4.1 Flash successor.26Multimodal agent modelsDeepSeek V4.1 FlashDeepSeek's multimodal Flash model, updated to V4.1-Flash-1017 on October 17, with the same `deepseek-flash` API ID, 1M context and peak/off-peak prices.27AI modelsGLM 5.2GLM 5.2 is Z.AI’s earlier open-weight language model for long-horizon coding, 1M context and tool-based workflows; use provider-specific access and retirement terms.28AI modelsHolo 3.1H Company computer-use vision-language model family for web, desktop, mobile, hosted API, and local agent deployments.29Reasoning modelsMAI-Thinking-1Microsoft AI reasoning model announced on June 2, 2026, with a 35B-active sparse MoE footprint, 256k context, coding and math benchmark claims, and private preview availability in Microsoft Foundry.30Coding modelsMAI-Code-1-FlashMicrosoft AI coding model announced on June 2, 2026, described as a 5B-parameter, inference-efficient agentic coding model integrated into GitHub Copilot, VS Code, and the Microsoft stack.31Speech modelsMisoTTSOpen-weight 8B text-to-speech model from Miso Labs for expressive speech and dialogue generation with text plus optional audio context.32Speech modelsMOSS-TTSOpen-source speech and sound generation model family from MOSI.AI and OpenMOSS for voice cloning, long-form TTS, dialogue speech, real-time streaming, voice design, sound effects, and local deployment.33Document AI modelsUnlimited OCRBaidu OCR model and code release for one-shot long-horizon document parsing.34Coding and agent modelsGrok 4.5xAI coding and agent model trained with Cursor for long-running software, data, office, and tool-using tasks.35Multimodal coding and agent modelsKimi K3Moonshot AI flagship multimodal model with 2.8 trillion parameters, a 1M-token context window, and long-horizon coding and knowledge-work capabilities.36Voice modelsGPT-Live-1OpenAI ChatGPT Voice model for full-duplex spoken conversations, rolling out to consumer ChatGPT plans.37Agentic multimodal modelsMuse Spark 1.3Meta multimodal reasoning model for coding, computer use, tool use, and long agentic workflows.38Open agentic coding modelsLaguna S 2.1Poolside open-weight 118B-A8B Mixture-of-Experts model for agentic coding, long-horizon work, and million-token contexts.39Image generation and editing modelsQwen-Image 3.0Third-generation Qwen image model focused on information-dense layouts, small text, realistic detail, multilingual rendering, and image editing.40On-device vision-language modelsVisionPsy-NanoQVAC family of compact open-weight vision-language models for single-image understanding on phones and edge devices.41Compact open-weight language modelsNeutrino-1Fermion Research family of ternary-format language models designed to run from one artifact on datacenter GPUs, laptops, and desktop CPUs.42Multimodal reasoning and agent modelsQwen 3.8Alibaba Qwen model family spanning the hosted `qwen3.8-max`, the open 2.4T text checkpoint, the dense multimodal `Qwen3.8-27B`, and the open `Qwen3.8-Flash-Next` preview of Qwen4 architecture.43AI video generation modelsSeedance 2.5Live ByteDance Dreamina video model for longer, reference-controlled, editable 4K video workflows, while public API details remain incomplete.44Speech recognition modelsQwen3-ASR-FlashQwenCloud multilingual speech-recognition model with a stable DashScope model ID and per-second pricing.45Open-weight language modelsK-EXAONE 2.0LG AI Research mixture-of-experts family led by a 750B-total, 37B-active Apache 2.0 checkpoint.46Open-weight language modelsopenPangu 2.0 ProHuawei 505B-total, 18B-active MoE model with 512K context and an Ascend-oriented deployment path.47Multimodal AI modelsDeepSeek V4 Flash Vision ExpRetired experimental DeepSeek vision endpoint whose legacy API ID temporarily routes to V4.1 Flash.48Frontier agent modelsGPT-6 AstraOpenAI flagship for coding, editable 3D scenes, research and computer-use workflows, with plan- and product-specific access.49Multimodal agent modelsGemini 3.8 FlashGoogle GA Flash model for long-horizon software engineering, autonomous agents, and complex enterprise workflows.50Cybersecurity modelsGemini 3.8 Flash CyberGoogle cybersecurity-specialist model for trusted defenders, vulnerability discovery, and automated patching through the Fairwind Program.51Coding and agent modelsGrok 4.6SpaceXAI flagship model for long-running coding, agentic knowledge work, and ambitious interactive or visual projects.52Coding and agent modelsGPT-6 SolOpenAI GPT-6 workhorse for complex coding and agentic workflows at $2 input and $10 output per million tokens.53Efficient language modelsGPT-6 LunaOpenAI GPT-6 efficiency model for focused, high-volume work at $0.10 input and $0.50 output per million tokens.54Coding and agent modelsClaude Opus 5.5Anthropic model for long-running agentic coding and knowledge work with lower prices than Opus 5.55Coding and agent modelsGrok 4.7SpaceXAI frontier model for coding, agentic tasks, knowledge work, and the Grok Bot harness.56Speech recognition modelsGrok Voice Transcribe 2.0SpaceXAI speech-to-text model for batch files and live streaming with timestamps, diarization, and multichannel audio.57Coding and agent modelsClaude Sonnet 5.5Anthropic's current Sonnet model for everyday coding and agent work, with a 1M context window and $2/$10 API token rates.58Image generation modelsNano Banana 2.1Google’s stable Gemini image model for generation and conversational editing, with 1K, 2K and 4K output.59Multimodal reasoning modelsMistral Large 4Mistral Large 4 is a multimodal open-weight model with 1M context, hosted API access and released FP8/NVFP4 checkpoints.60Multimodal embedding modelsEmbeddingGemma 2Google’s open 740M model for local text, image, video and audio embeddings.61AI modelsClaude Haiku 5.5Anthropic's fast model for classification, extraction and subagent work, with prompt-length-based pricing.62AI modelsMuse WonderMeta's October 12 model for long-running agents, with API access distinct from the subscription-only Full variant.63AI modelsMuse Wonder FullMeta's higher-capability Wonder variant for Muse Code Power Usage and Maximum Usage subscribers.64Image generation and editingQwen Image 2.1 TurboQwen's October 9 eight-step image checkpoint, with transparent-image editing and research-only weight terms.65Visual reasoningElorian GlimpseElorian’s research-preview model for image and video reasoning, with API access, official demonstrations and provider-specific limits.66Open-weight multimodal modelsMuse Spark OpenMeta's 816B open-weight model for coding and tool workflows, with conditional commercial terms and substantial self-hosting requirements.67Language modelsReflection BeamReflection’s October 16 open-weight text reasoning and coding model, with Apache 2.0 weights in BF16, FP8 and NVFP4.68Language modelsNVIDIA Nemotron 3.5A model family with Super VL for large multimodal workloads and Bolt for a smaller packed local deployment.69Language modelsStep 5 PreviewStepFun’s sparse multimodal agent model, with downloadable FP8/BF16 checkpoints and conditional commercial licensing.70Language modelsGLM-LegalZ.ai’s legal-domain model built from GLM-5.3-Flash, with a beta API application and publisher workflow examples.71Visual reasoning and computer useGPT-6.1 TerraOpenAI model for visual analysis, computer use and writing with lower standard token prices than GPT-6.1 Sol.72Multimodal coding and agent modelsKimi K3.1 ProMoonshot’s post-trained K3 successor, with separate standard and Highspeed APIs, Kimi Code plan gates and a dated full-weight promise.73AI coding modelsClaude Fable 5.2Compare Fable 5.2 with Opus 5.5 and Fable 5.1, and check pricing, access, migration changes and evaluation limits before switching.
Entity type
Tools
01AI agent toolsREA (Reverse Engineer Anything)REA connects coding agents to software-analysis tools. Explore its creators, game reconstruction, the Adobe replacement story, and what AI reverse engineering can actually prove.02AI agent toolsESP32 iPhone AgentAn open-source iPhone agent with AirPlay eyes, ESP32-S3 USB mouse and keyboard input, and a vision model running off the board.03AI coding toolsClaude CodeAnthropic agentic coding tool that runs in the terminal, edits code, runs commands, and supports permissions, hooks, subagents, and enterprise controls.04AI coding toolsPi Coding AgentOpen-source terminal coding agent and minimal agent harness designed around extensions, skills, provider choice, and embeddable modes.05AI coding toolsOpenCodeOpen-source AI coding agent for terminal, desktop, and editor-adjacent workflows, maintained in the anomalyco OpenCode repository.06AI coding toolsZerostackMinimal Rust coding agent optimized for memory footprint, performance, provider selection, prompts, and optional sandboxed command execution.07AI coding toolsDeepSeek ReasonixDeepSeek-native terminal coding agent designed around prefix-cache stability, tool-call repair, MCP, skills, memory, and low-cost long sessions.08AI coding toolsDevSpaceSelf-hosted MCP server that gives ChatGPT or another MCP host controlled access to local coding workspaces.09AI coding toolsZotOne-binary Go coding agent harness with built-in tools, provider routing, extensions, sessions, and background subagents.10AI coding toolsSmallCodeOpen-source terminal coding agent optimized for small local language models and constrained context windows.11Model and inference infrastructureDeepSpecDeepSeek open-source codebase for training and evaluating speculative decoding algorithms.12AI coding toolsZCodeZ.ai desktop coding harness optimized for GLM-5.2 agentic engineering workflows.13AI agent toolsAgent-ReachOpen-source tool package that gives AI agents read-oriented access to web and social content sources.14AI agent infrastructureTabstackMozilla-backed web execution, extraction, research, and browser automation API for AI agents.15AI agent infrastructureHumalikeBehavioral API layer for humanlike AI agents, group conversation, turn-taking, social memory, and social observability.16AI agent toolsPage AgentAlibaba open-source JavaScript in-page GUI agent that lets developers add natural-language web control to their own applications.17AI agent toolsMicrosoft FlintMicrosoft open-source visualization language and MCP server for agent-generated charts.18Coding agent memoryDeja VuA local search and recall tool for coding-agent session histories across supported harnesses.
Entity type
Products
01Personal AI agentsOpenAI DotsOpenAI's proactive personal-agent product uses GPT-6 Astra, a cloud computer, and approved connected apps.02Personal AI agentsCue by ManusCue is Manus's separate personal-agent app, designed around agents with their own phone, email, wallet, and computer.03AI agent workspacesManus 2.0Manus 2.0 connects a new agent harness, a hands-on Studio workspace, event-triggered automation, computer use, and persistent project environments.04Cloud AI agentsGrok BotPersistent AI teammates that use a shared cloud computer to work across apps and run recurring jobs.05AI design toolsClaude DesignAnthropic Labs product for creating designs, prototypes, presentations, one-pagers, and other visual work with Claude.06Personal AI agentsMuseMeta personal AI agent that runs on a dedicated cloud computer, works through its own app or WhatsApp, and continues approved tasks in the background.07Image generation and editingChatGPT Images 2.6OpenAI’s November 2026 image release adds GIF creation, SVG output and more editing controls, with separate Flare and Sunburst API models.08Image generation and editingChatGPT Images 2.5OpenAI image-generation release spanning the ChatGPT creation experience and the GPT-Image-2.5 Flare and Sunburst API models.09Agent infrastructureRuntimeTeam agent runtime for sandboxed coding agents with context, integrations, guardrails, observability, orchestration, and self-hosting options.10AI coding toolsSupersetOpen-source AI coding platform for running CLI agents in parallel across isolated git worktrees, with terminal, review, diff, IDE, and MCP surfaces.11Agent infrastructureInsForgeOpen-source backend platform for agentic coding, giving coding agents database, auth, storage, compute, hosting, deployment, and an AI model gateway.12AI agent infrastructureMinicorRPA platform for deploying computer-use agents into legacy desktop and web systems with self-healing, observability, and Windows VM execution.13AI model systemsN71Memory-research company developing an evidence-anchored context and recall engine for AI systems.14AI billing infrastructureStigg 2.0Usage runtime for AI products that checks credits, entitlements, budgets, and metering before expensive AI work runs.15Agentic paymentsSequence AgenticMoney-movement API for AI agents with scoped permissions, pre-approved rules, audit trails, and idempotent transfers.16Local AI assistantsJan AIAn open-source local AI assistant for running and connecting models on your own computer.17Open-model desktop agentsLM Studio BionicSeparate LM Studio desktop agent for coding, research, documents, files, local models, and optional secure-cloud inference.18Multimodal creative agentsMiniMax DesignFormerly MiniMax Hub, this desktop creative-agent workspace coordinates text, image, video, and audio work on a shared canvas.19AI infrastructureMoondreamOpen vision-model project acquired by Prime Intellect; Moondream Cloud inference and Lens are scheduled to close November 13, 2026.20Proactive AI assistantsClaude DuetAnthropic’s cloud assistant checks connected work, prepares tasks and asks for consequential-action approval, within a US Pro/Max waitlist preview.21Local desktop AI assistantsHobFrigade’s MIT desktop coworker selects local open-weight models for files, PDFs, charts and code, with optional network access.
Entity type
Companies
01Brain-computer interfacesNeuralinkNeurotechnology company developing the N1 implanted brain-computer interface and participant-specific assistive control systems.02Open AI infrastructureHugging FaceThe open AI platform whose announced NVIDIA acquisition price hides a Unicode easter egg.03AI companiesDeepSeekChinese AI company behind DeepSeek Chat, model APIs, open-weight releases, and the multimodal V4.1 Flash architecture.
Entity type
Features
01AI coding toolsCodex PetsAnimated companion overlays for the Codex app.02AI coding toolsClaude BuddyThe removed Claude Code April Fools companion, with its history and current alternatives.03AI agentsGemini SparkGoogle Gemini app personal AI agent for background tasks across Google tools, MCP-connected services, and future desktop or browser workflows.04AI agentsClaude TagAnthropic Slack-native agent feature that lets teams tag Claude into shared threads, tools, and longer-running work.05In-car AI assistantsTesla Doubao Voice AssistantTesla China conversational beta assistant, with AMD hardware, software and connectivity eligibility checks.
Entity type
Frameworks
01Personal AI agentsOpenClawAn open-source personal assistant built around a user-controlled Gateway, with messaging channels, searchable memory, skills and scheduled jobs.02Personal AI agentsHermes AgentNous Research’s open-source personal agent with persistent memory, learned skills, Bot Mode, messaging gateways and scheduled work.03Web 3D frameworksthree.jsJavaScript 3D framework for rendering interactive browser scenes with WebGL, WebGPU, WebXR, and common 3D asset formats.04Agent infrastructureForgePython reliability layer for self-hosted LLM tool-calling, guardrails, context management, backend adapters, and multi-step agentic workflows.05AI design infrastructureDESIGN.mdGoogle Labs format specification for giving coding agents persistent design-system context.06Model serving infrastructureDeepInferDeepSeek’s Apache-2.0 offline inference engine exposes optimized kernels and distributed execution, while complete online service components remain separate.
Entity type