Model Context Protocol (MCP) Server Directory

Inkling (Thinking Machines Lab, 975B MoE, Apache 2.0) vs GLM 5.2 (Zhipu AI, top coding/reasoning) vs Nemotron 3 Ultra (NVIDIA, 550B): benchmark comparison across SWEBench, Terminal Bench, MCP Atlas, HLE, AIME 2026, and c...

Mira (Clockwork.ai, launched July 8, 2026) is the first Agentic FP&A analyst for small businesses. It connects to QuickBooks or Xero in 3 minutes and automatically generates cash flow forecasts, scenario models, balance...

Meetily is a privacy-first, MIT-licensed AI meeting assistant with 22,997 GitHub stars that runs 100% locally on the user's machine. Unlike Otter.ai ($8-30/month), Fireflies.ai ($10-29/month), and Granola ($18/month) whi...

OpenSquilla (5,552 GitHub stars, Apache 2.0) is a token-efficient microkernel AI agent with an on-device LightGBM + ONNX classifier (SquillaRouter) that routes each turn to the cheapest capable model across four tiers (C...

Frugon (4,200+ GitHub stars, July 2026) is an open-source intelligent model routing engine that sends simple queries to cheap models and complex ones to frontier models. Portkey is a managed AI gateway with fallback stra...

NVIDIA Audex 30B-A3B (July 2026) and Qwen3.5-35B-A3B are the two leading open audio-text LLMs. Audex uniquely handles both audio understanding and generation in a single model while preserving text intelligence. Qwen3.5-...

Meta Muse Image (July 2026) is Meta's first image generation model from Superintelligence Labs, available free in Meta AI, Instagram Stories, and WhatsApp with @-mention tagging, sketch-based editing, and Advantage+ ad i...

Hy3 beats GLM-5.2 on search and tool orchestration (84.2 BrowseComp, 79.1 MCP-Atlas) with Apache 2.0 license and single-node deployment. GLM-5.2 leads on coding (84.2 SWE-bench Verified) but requires 8x H200 nodes. Hy3 w...

Otari is Mozilla.ai's open-source LLM control plane with virtual keys, budgets, and built-in tools (code execution, web search). Portkey is an enterprise AI gateway with observability, guardrails, and managed hosting. Li...

Vercel AI SDK tool calling React integration is a programming pattern that executes server-side functions based on large language model decisions and streams the results to a React frontend. By combining streamText with...

Tavily vs Firecrawl comparison evaluates two web data acquisition services optimized for retrieval-augmented generation pipelines. Selecting the correct service reduces average web context extraction latency from nine an...

Mem0 vs LangChain memory hybrid system combines Mem0 v0.2.0 semantic fact extraction with LangChain RunnableWithMessageHistory session tracking to manage persistent user context. The system runs both components in parall...