Ollama
Run open LLMs locally with a single command.
Open-source infrastructure and hosted MCP servers with built-in OAuth across 600+ tools.
Klavis AI hits a nice sweet spot: open-source so you can self-host and audit, but with a managed option that handles the genuinely hard parts (multi-tenant OAuth across hundreds of tools). The Slack/Discord/web clients and broad SDK support are practical touches for shipping quickly. Caveats: it's a young, YC-stage company, so expect rapid change and evaluate production reliability yourself; hosted-tier limits and enterprise pricing need direct confirmation. Verify current coverage and pricing with the vendor.
Klavis AI is an open-source, YC-backed platform providing hosted, secure MCP servers with built-in OAuth and multi-tenant auth across 600+ tools and SaaS apps. It offers MCP clients for Slack, Discord, and the web, plus Python/TypeScript SDKs and a REST API. Teams can self-host the open-source core or use the managed service.
Klavis AI, founded by ex-Google DeepMind and ex-Lyft engineers and backed by Y Combinator (X25), provides open-source infrastructure to build, use, and scale MCP servers for AI applications. Its hosted, secure MCP servers eliminate the need to manage authentication and client code, with built-in OAuth and multi-tenant auth spanning 600+ tools and SaaS apps, including production-ready servers for GitHub, Slack, Gmail, Salesforce, Linear, and Notion. Beyond servers, Klavis offers MCP clients for Slack, Discord, and the web so users can invoke MCP tools inside those platforms, plus Python/TypeScript SDKs and a REST API with support for OpenAI, Claude, Gemini, LangChain, LlamaIndex, and CrewAI. Because the core is open source, teams can self-host, or use the managed service for hosted servers and enterprise OAuth. As with all MCP infrastructure in 2026, coverage and pricing evolve quickly.
Klavis AI is a YC-backed, open-source MCP infrastructure platform offering hosted servers with built-in OAuth across 600+ tools, plus Slack/Discord clients and SDKs.
Klavis AI is a Y Combinator (X25) startup founded by ex-Google DeepMind and ex-Lyft engineers, building open-source infrastructure for MCP.
It targets developers who want to use, build, and scale MCP servers without managing authentication and client code.
Klavis offers hosted, secure MCP servers with built-in OAuth and multi-tenant auth across 600+ tools, MCP clients for Slack, Discord, and the web, and Python/TypeScript SDKs plus a REST API.
The open-source core can be self-hosted, and the platform supports major LLM providers and agent frameworks.
Developers and teams building agentic applications that need authenticated tool access across many SaaS apps, with the option to self-host.
Developers wiring authenticated tools into agents.
Engineering leads at startups building agent products.
YC network, open-source and MCP communities.
A startup or team building multi-tenant agentic products that wants open-source flexibility plus managed OAuth across many SaaS tools.
Klavis AI is backed by Y Combinator (X25); verify any additional funding, which may have changed since.
Yes. Its core infrastructure is open source and self-hostable, with an optional managed service for hosted servers and OAuth.
Klavis runs managed MCP servers with built-in OAuth across 600+ tools and SaaS apps.
Yes. Klavis offers MCP clients for Slack, Discord, and the web.
It was founded by ex-Google DeepMind and ex-Lyft engineers and is backed by Y Combinator (X25).
Via Python/TypeScript SDKs or a REST API, with support for OpenAI, Claude, Gemini, LangChain, LlamaIndex, and CrewAI.
Side-by-side pages for pricing, features, and best-fit use cases.
Run open LLMs locally with a single command.
The open hub for machine learning models, datasets, and demos.
Open-weight frontier LLM family from Z.ai (Zhipu AI), tuned for coding and agents.
Alibaba's free AI assistant, backed by the open-weight Qwen model family