Trust scores for
the entire AI stack

Anthropic's top-tier model above Opus and the most capable widely released Mythos-class model. State-of-the-art on nearly all tested benchma...

SUPERSEDED: no longer Anthropic's most capable model — succeeded by Opus 4.6, 4.7, 4.8 (2026-05-28) and Claude Fable 5 (2026-06-09, new top ...

Anthropic's frontier Opus released February 2026 with 80.8% SWE-bench Verified, breakthrough 68.8% ARC-AGI-2 abstract reasoning, adaptive th...

Anthropic's flagship Opus model with state-of-the-art long-horizon agentic execution, knowledge work, and memory. 84% on Online-Mind2Web, dy...

OpenAI's first open-weight model released August 2025. 117B total params (5.1B active), Apache 2.0 license. Matches o4-mini on many benchmar...

OpenAI's edge-optimized open-weight model released August 2025. 21B total params (3.6B active), Apache 2.0 license. Matches o3-mini despite ...

Official MCP reference server for time and timezone operations. Provides AI models with current time information, timezone conversions, date...

DEPRECATED: Anthropic deprecated Claude Opus 4.1 (claude-opus-4-1-20250805) on 2026-06-05; it will be retired on 2026-08-05 and requests wil...

Previous-generation Opus flagship, superseded by Opus 4.8. 64.3% SWE-Bench Pro and 94.2% GPQA Diamond at launch. First Claude with high-reso...

Anthropic's previous-generation Sonnet workhorse, superseded by Claude Sonnet 5 (2026-06-30) as the best speed/intelligence balance at the s...

Anthropic's Sonnet-tier flagship (released 2026-06-30), positioned as the cheaper way to run agents: 72.7% SWE-bench Verified (vs Sonnet 4.6...

Google's current flagship reasoning model with 77.1% ARC-AGI-2 (2.5x Gemini 3 Pro), 94.3% GPQA Diamond, 2887 Elo on LiveCodeBench Pro, and 1...

SUPERSEDED by GPT-5.4 (2026-03-05), GPT-5.5 (2026-04-23), and the GPT-5.6 family (2026-07-09). API snapshots (gpt-5.2, gpt-5.2-2025-12-11) s...

OpenAI's previous-generation flagship (superseded by GPT-5.5 in April 2026 and the GPT-5.6 family in July 2026). Fully supported with no ann...

OpenAI's flagship from April 2026 (codename 'Spud'), first fully retrained base model since GPT-4.5. Succeeded at the top of the lineup by t...

Anthropic's fastest model, released October 2025 and still the speed tier of the current lineup (Active as of 2026-07). At launch it beat th...

Previous-generation Sonnet released September 2025, since superseded by Sonnet 4.6 (2026) and Claude Sonnet 5 (2026-06-30). Still Active on ...

SHUT DOWN: Google retired Gemini 3 Pro Preview on 2026-03-09 (the gemini-pro-latest alias moved to 3.1 Pro on 2026-03-06); it is no longer s...

DEPRECATED: superseded by GPT-5.3-Codex (2026-02-05); gpt-5.2-codex API shuts down 2026-07-23 (two weeks from 2026-07-09 — migrate now; Open...

Fully managed AWS service for building and deploying generative AI agents with orchestration, memory, knowledge bases, and action groups. No...

RETIRED: Anthropic retired Claude Opus 4 (claude-opus-4-20250514) on 2026-06-15 (deprecated 2026-04-14); API requests now fail. Recommended ...

DEPRECATED: Google has scheduled Gemini 2.5 Pro for shutdown on 2026-10-16; the designated replacement is Gemini 3.1 Pro. Formerly Google's ...

Google's GA 'frontier workhorse' launched at I/O 2026. Beats Gemini 3.1 Pro on agentic and coding suites (76.2% Terminal-Bench 2.1, 83.6% MC...

Google's efficiency model with Pro-level performance at low cost. 78% SWE-bench (beat Gemini 3 Pro), 1M context, 3x faster than 2.5 Pro. Thi...

SUPERSEDED by GPT-5.2/5.4/5.5 and the GPT-5.6 family (released 2026-07-09); retired from ChatGPT 2026-03-11. gpt-5.1-chat-latest, gpt-5.1-co...

OpenAI's agentic coding specialist, still active with no announced shutdown (as of 2026-07-09): ~80% SWE-bench Verified, 77.3% Terminal-Benc...

OpenAI's GPT-5.6 family — Sol (flagship, $5/$30), Terra (balanced, $2.50/$15), Luna (fast, $1/$6 per 1M) — publicly released 2026-07-09. Day...

DEPRECATED: retired from ChatGPT (GPT-5 Instant/Thinking removed by 2026-02-13); gpt-5-chat-latest and gpt-5-codex API shut down 2026-07-23;...

Meta's flagship open-weight model (released April 5, 2025): a mixture-of-experts with 400B total / 17B active parameters (128 experts), nati...

DEPRECATED: o1 variants and o1-pro shut down in the API on 2026-10-23; already removed from ChatGPT (o1-preview/o1-mini removed in 2025). Mi...

AWS managed conversational AI service for building chatbots and voice assistants with automatic speech recognition (ASR) and natural languag...

REBRANDED: at Cloud Next (April 2026) Vertex AI Agent Builder became the Gemini Enterprise Agent Platform (APIs unchanged), following Agents...

MAINTENANCE MODE: Semantic Kernel now receives bug/security fixes only and is superseded by the Microsoft Agent Framework (1.0 GA on 2026-04...

RETIRED: Anthropic retired Claude Sonnet 4 (claude-sonnet-4-20250514) on 2026-06-15 (deprecated 2026-04-14); API requests now fail. Recommen...

LEGACY/RETIRING: the underlying Bot Framework SDK is retired — final long-term support ended December 31, 2025 (no further updates; Azure po...

Enterprise 'Work AI' platform that unifies information across business tools and applications. Has expanded well beyond enterprise search in...

REBRANDED: Dialogflow CX is now sold as Conversational Agents on Google Cloud - the standalone CX console was deprecated October 31, 2025, a...

Cohere's Apache 2.0 open-weight 218B sparse MoE (25B active) unifying Command A, A Reasoning, A Vision, and A Translate. Runs on 2xH100 or a...

SHUT DOWN: Google shut down the Gemini 2.0 Flash family on 2026-06-01; the model is no longer served via the Gemini API. Historically a fast...

Meta's largest open-source model with 405 billion parameters, offering complete transparency, self-hosting capabilities, and competitive per...

253B parameter model from NVIDIA's Llama-3.1-based Nemotron line, now superseded: NVIDIA discontinued this line in favor of the native Nemot...

Amazon's cost-efficient Nova 2 workhorse model, GA on Amazon Bedrock since re:Invent 2025. Offers three thinking-intensity levels, a built-i...

DEPRECATED: o3-mini's API shuts down 2026-10-23; migration target is GPT-5.5. Historically an efficient reasoning model from OpenAI achievin...

DEPRECATED: the entire o3 family now has shutdown dates — o3 (o3-2025-04-16) and o3-pro shut down in the API on 2026-12-11 (announced 2026-0...

DEPRECATED: removed from ChatGPT 2026-02-13; o4-mini API shuts down 2026-10-23 (o4-mini-deep-research shuts down 2026-07-23); OpenAI's recom...

CONSOLIDATED: Watson Assistant was rebranded watsonx Assistant (2023) and folded into IBM watsonx Orchestrate - the Assistant pricing page n...

Enterprise agentic AI platform for designing, deploying, managing, and scaling AI agents across business operations. In May 2026 Kore.ai lau...

Visual automation platform with AI capabilities for building no-code intelligent workflows. Integrates AI models with 3,000+ app connections...

TRANSITIONING: Einstein Copilot was retired into Agentforce (Jan 2025, Spring '25 release) and Salesforce is steering Einstein Bots customer...

SUPERSEDED by the Nova 2 family announced at re:Invent 2025-12-02 (Nova 2 Lite GA; Nova 2 Pro/Omni still in preview as of 2026-07-09), thoug...

No-code automation platform with AI actions for connecting AI models to 8,000+ apps. Enables building intelligent automations and AI-powered...

LEGACY: retired from ChatGPT 2026-02-13 but still available in the API with no announced shutdown (as of 2026-07-09). Previous-generation ge...

Meta's powerful 70B parameter Llama 3.3 model offering strong performance with open-source flexibility and an excellent balance of capabilit...

Meta's announced 2T-total/288B-active parameter Llama 4 teacher model that was NEVER RELEASED. It remains 'announced, not released' as of Ju...

Meta's efficient Llama 4 model (released April 5, 2025): a natively multimodal mixture-of-experts with 109B total / 17B active parameters (1...

Mistral AI's open-weight flagship released December 2025 under Apache 2.0: a sparse MoE (675B total / 41B active) multimodal model with ~256...

NVIDIA's open frontier reasoning model, released 2026-06-04 to complete the Nemotron 3 rollout (Nano Dec 2025, Super Mar 2026): a 550B total...

Alibaba's Apache-2.0 flagship open model: Qwen3.5-397B-A17B, a 512-expert hybrid MoE (397B total / 17B active), natively multimodal, 262K co...

MAINTENANCE MODE: AutoGen now receives bug/security fixes only, is community-managed going forward, and is superseded by the Microsoft Agent...

DEPRECATED: the Assistants API will be sunset on 2026-08-26 (under 7 weeks away); after that date all /v1/assistants, /v1/threads, and /v1/t...

DeepSeek's standalone reasoning model, now superseded and discontinued. Its reasoning was folded into DeepSeek V3.1's hybrid thinking mode (...

DeepSeek's ~685B-parameter MoE flagship with DeepSeek Sparse Attention (DSA) for dramatically cheaper long-context inference. The V3.2-Speci...

DEPRECATED: removed from ChatGPT 2026-02-13 and fully retired from ChatGPT (including Custom GPTs) 2026-04-03; chatgpt-4o-latest API access ...

RETIRED: o1-mini was shut down in the OpenAI API on 2025-10-27 (o1-preview on 2025-07-28) and removed from ChatGPT; it is no longer availabl...

Alibaba's Apache-2.0 open-weight Qwen3.6 family (Apr 2026): Qwen3.6-35B-A3B MoE (35B total / 3B active, 2026-04-16) and Qwen3.6-27B dense (2...

Open-source SDK and runtime for building AI agents and graph-based multi-agent workflows in .NET and Python. Merges AutoGen and Semantic Ker...

Production-ready multi-agent orchestration framework built around agents, handoffs, guardrails, and tracing. Open-source (MIT) successor to ...

Type-safe Python agent framework from the creators of Pydantic. Reached v1.0 stable on 2025-09-04 with a formal API stability commitment, th...

Open-source, model-driven AI agent SDK from AWS, used internally by Amazon Q Developer. Takes a lightweight model-first approach with MCP an...

Perplexity's official MCP server (perplexityai/modelcontextprotocol, npm @perplexity-ai/mcp-server) for the Perplexity API Platform. Exposes...

Stripe's official MCP server (repo renamed from stripe/agent-toolkit to stripe/ai). The hosted server at mcp.stripe.com (OAuth or restricted...

Official Supabase MCP server: SQL execution, schema/migrations, type generation, Edge Function deploys, logs, storage, branching. Primarily ...

DeepSeek's preview flagship family: V4-Pro (1.6T total / 49B active MoE, largest open-weight release ever) and V4-Flash (284B/13B). 1M conte...

LEGACY: retired from ChatGPT 2026-02-13 but still available in the API with no announced shutdown (as of 2026-07-09). Balanced GPT-4.1 varia...

RETIRED: xAI retired Grok 3 on 2026-05-15; retired API slugs now silently redirect to Grok 4.3 at Grok 4.3 pricing. Historically xAI's flags...

xAI's late-2025 flagship that debuted #1 on LMArena Text (1483 Elo) and led EQ-Bench3 for emotional intelligence, with a 2M token context wi...

xAI's workhorse model (released 2026-04-30): 1M context, reasoning, function calling, and structured outputs at $1.25/$2.50 per 1M tokens. S...

Secure cloud runtime for AI agents with code interpreter capabilities. Provides sandboxed environments for executing agent-generated code sa...

Open-source, code-first framework for building, evaluating, and deploying AI agents. Supports workflow agents, multi-agent hierarchies, buil...

LangChain's graph-based agent framework for building stateful, multi-actor applications with cycles and controllable execution flow. Reached...

Fair-code workflow automation platform with AI agent capabilities, now on the 2.x major version. Visual builder with AI agent nodes and 400+...

Conversational AI platform for autonomous customer experience agents, built on Sierra's Agent OS 2.0 spanning chat, voice, email, SMS, and W...

ARCHIVED: Former Anthropic reference MCP server for the Brave Search API, archived 2025-05-29 to the servers-archived repository and no long...

Official MCP reference server enabling dynamic, extended reasoning and problem-solving sequences: structured thinking processes, decompositi...

Google's open-source Gemma 3 model with 27 billion parameters, now superseded by Gemma 4 (released 2026-04-02 under Apache 2.0, a license im...

Z.ai's MIT-licensed 744B-parameter MoE (40B active) launched June 2026 with a 1M-token context via IndexShare sparse attention. Leading open...

Z.ai's MIT-licensed 744B-parameter MoE (40B active) with 77.8% SWE-bench Verified, 92.7% AIME 2026, and open-source leadership on BrowseComp...

SpaceXAI's flagship (released 2026-07-08, days after the xAI-to-SpaceXAI rebrand): an 'Opus-class' model tuned for token efficiency at $2/$6...

Multimodal vision-language model from Alibaba, now three generations behind: superseded by Qwen3-VL (Sep 2025), the natively-multimodal Qwen...

Open-source LLM application and agentic workflow platform with a visual canvas, built-in RAG pipeline, agent nodes, and 50+ tools. One of th...

Open-source AI orchestration framework from deepset for building production-ready LLM applications, RAG pipelines, agent workflows, and sema...

Data framework optimized for building LLM applications with advanced RAG (Retrieval-Augmented Generation) capabilities. Agents can reason ov...

OpenAI's coding agent spanning a cloud agent that runs tasks in isolated containers and an open-source CLI. Delegates parallel software task...

Open-source-rooted conversational AI framework for building contextual assistants with full control over NLU, dialogue management, and deplo...

Exa's official MCP server (exa-labs/exa-mcp-server v3.2.1, MIT) exposing neural/semantic web search, code search, crawling/content fetch, co...

GitHub's OFFICIAL MCP server, successor to the archived Anthropic reference server. Open source (Go, MIT); binary/Docker or hosted remote at...

Official Grafana Labs MCP server (grafana/mcp-grafana, Apache-2.0, Go) with 40+ tools: query metrics/logs/traces across Prometheus, Loki, an...

MCP server built into the shadcn CLI (run via npx shadcn@latest mcp) that lets AI agents browse, search, and install UI components from any ...

Google's open-weight family released April 2026 under Apache 2.0 (a shift from the custom Gemma license). Spans E2B/E4B edge models with 128...

OpenAI's efficient multimodal model combining text and vision capabilities at competitive pricing. Designed for cost-sensitive applications ...

Moonshot AI's open-weight 1T-parameter MoE (32B active) with vendor-reported 80.2% SWE-Bench Verified and 58.6 SWE-Bench Pro. Agent Swarm or...

TypeScript-first AI agent framework from the Gatsby founders, combining agents, durable workflows, RAG, and evals in one toolkit. Provider-a...

Databricks managed MCP servers: workspace-hosted endpoints under https://<workspace>/api/2.0/mcp/ for Genie natural-language data queries, A...

Official MCP reference server providing AI models with controlled access to the local filesystem (file reading, writing, and directory opera...

ARCHIVED: Former Anthropic reference MCP server for GitLab (merge requests, CI/CD pipelines, issues), archived 2025-05-29 to servers-archive...

Official Snowflake MCP server, GA since 2025-11-04 as a fully managed in-account endpoint (CREATE MCP SERVER object). Exposes Cortex Analyst...

Tavily's official MCP server (tavily-ai/tavily-mcp, MIT) enabling AI models to perform real-time web search, content extraction, site mappin...

DEPRECATED: OpenAI announced 2026-04-22 that gpt-4.1-nano's API shuts down 2026-10-23; recommended replacement is gpt-5.4-nano. Historically...

Moonshot AI's coding-specialized open-weight MoE (1T total / 32B active, Modified MIT) built on Kimi K2.6, released 2026-06-12. Vendor repor...

MiniMax's MIT-licensed 230B MoE with only 10B active parameters, optimized for agentic tool calling and coding. Topped open-model agentic ra...

Anthropic's agentic coding tool available as a terminal CLI, IDE extensions, web, and desktop app. Plans and executes multi-step coding task...

No-code AI agent platform that enables teams to build, customize, and deploy AI workforce agents. Features visual workflow builders, tool in...

Official Cloudflare catalog of managed remote MCP servers, connected over OAuth. The primary Cloudflare API server at https://mcp.cloudflare...

Official Datadog MCP server for observability integration, generally available since March 2026 as a hosted streamable-HTTP remote (US1: htt...

Official MCP reference server for fetching web content and converting HTML to markdown. Enables AI models to retrieve and process web pages,...

Official MCP reference server for local Git operations (commits, branches, version control). One of the seven reference servers still mainta...

Microsoft's official MCP server for browser automation via Playwright. Exposes 25+ tools (navigation, clicking, typing, form filling, screen...

DeepSeek's March 2025 open-weights V3 checkpoint, now superseded by DeepSeek V3.1 (Aug 2025), V3.2 (Dec 2025), and V4 (Apr 2026). The legacy...

Open-source no-code business automation platform with AI agent capabilities. Self-hostable alternative to Zapier with visual workflow builde...

SDK exposing Claude Code's production agent harness (tool loop, permission system, subagents, MCP) for building general-purpose agents in Ty...

Role-playing multi-agent framework for orchestrating collaborative autonomous agents. Agents work as a crew with defined roles, goals, and b...

Minimalist Python agent library from Hugging Face. Its signature CodeAgent writes actions as executable Python code instead of JSON tool cal...

Official Atlassian (Rovo) MCP server, generally available since February 2026 as a hosted remote at https://mcp.atlassian.com/v1/mcp (OAuth ...

Official Microsoft MCP server for Azure cloud services, generally available as Azure MCP Server 2.0 (2026-04-10) with tools spanning 44+ Azu...

Official ClickHouse MCP family: open-source mcp-clickhouse (PyPI, v0.4.0) plus ClickHouse Cloud Remote MCP at https://mcp.clickhouse.cloud/m...

Upstash's documentation-retrieval MCP server. Two tools (resolve-library-id, get-library-docs) inject up-to-date, version-specific library d...

Official MCP reference server demonstrating all protocol features (tools, resources, prompts, sampling) for testing and development. One of ...

Official Firecrawl MCP server (firecrawl-mcp v3.x) providing web scraping, crawling, site mapping, web search, structured extraction, docume...

Linear's OFFICIAL MCP server, a hosted remote at https://mcp.linear.app/mcp using OAuth 2.1 with dynamic client registration; Bearer-token/A...

Official Neon MCP server for managing serverless Postgres via natural language: projects, branches, SQL execution, and branch-based safe mig...

PayPal's official MCP server (Agent Toolkit), hosted at https://mcp.paypal.com with SSE (/sse) and streamable HTTP (/http) transports plus a...

Salesforce's official MCP offering. Primary path is Hosted MCP Servers (GA 2026-04-29, included for Enterprise Edition+ and Developer Editio...

Official Sentry MCP server (getsentry/sentry-mcp) for error tracking and monitoring integration. Sentry hosts a managed remote at https://mc...

ARCHIVED WITH UNPATCHED VULNERABILITY: Former Anthropic reference MCP server for SQLite, archived 2025-05-29 to the servers-archived reposit...

Vercel's official hosted MCP server at mcp.vercel.com. Provides docs search plus tools to inspect teams, projects, deployments, build and ru...

Zapier's hosted, proprietary MCP server giving AI agents access to user-selected actions from 9,000+ connected apps (40,000+ actions). Users...

MiniMax's natively multimodal 428B MoE (23B active) with a 1M-token context via MiniMax Sparse Attention, launched June 2026 with weights on...

The most-adopted open-source coding agent (5M+ installs by Feb 2026), formerly 'Claude Dev'. Apache-2.0, runs in VS Code, JetBrains, termina...

Open-source, local-first AI agent written in Rust, created by Block and donated to the Linux Foundation's Agentic AI Foundation (repo moved ...

REBRANDED: MemGPT became Letta (Letta Inc., a UC Berkeley spinout) in September 2024; "MemGPT" now refers only to the research technique. Le...

Community-maintained MCP server for Google Calendar integration. Enables AI models to create, read, update, and delete calendar events, mana...

ARCHIVED: Former Anthropic reference MCP server (gdrive) for accessing and searching Google Drive files, archived 2025-05-29 to servers-arch...

MCP servers for AWS S3 storage operations, enabling AI models to upload, download, list, and manage objects in S3 buckets. The landscape has...

Open-source low-code platform for LLM orchestration flows and AI agents, acquired by Workday August 2025. Visual node editor for RAG pipelin...

Visual builder for LangChain-based AI apps and agents; owned by IBM via the Feb 2025 DataStax acquisition. SECURITY: recurring CVEs - CVE-20...

MIT-licensed open-source terminal AI coding agent from the SST team (Anomaly Innovations). TUI-first with a client/server architecture, 75+ ...

Warp's Agentic Development Environment (ADE): a Rust-based terminal reimagined for prompt-driven, multi-agent software development, paired w...

Official Apify MCP server (renamed from actors-mcp-server) exposing thousands of Apify Store Actors - scrapers for social media, maps, e-com...

Google's official MCP server that gives AI agents full Chrome control via the Chrome DevTools Protocol. Around 26 tools span input automatio...

Hugging Face's official MCP server connecting AI assistants to the Hub. Ships 7 built-in tools (search for models, datasets, Spaces, and pap...

Notion's official MCP integration. Primary offering is the hosted server at https://mcp.notion.com/mcp (Streamable HTTP and SSE, one-click O...

ARCHIVED: Former Anthropic reference MCP server for Redis cache and data structure operations, archived 2025-05-29 to the servers-archived r...

Shopify's official MCP surface spans three layers: the Storefront MCP, live by default on every eligible store at {shop}.myshopify.com/api/m...

Autonomous data labeling agent framework for creating self-improving AI systems. Combines LLMs with ground truth learning to automate and im...

Open-source terminal AI agent from Google that brings Gemini into the command line. Uses a ReAct loop with built-in tools, MCP server suppor...

AWS's official MCP offering for AWS cloud services. Two forms: the managed AWS MCP Server (GA 2026-05-06), a fully managed remote exposing 1...

Box's official MCP server, a hosted remote at https://mcp.box.com using OAuth 2.0 with admin-managed enablement (Admin Console > Integration...