Glama

aid: glama-ai
url: https://raw.githubusercontent.com/api-evangelist/glama-ai/refs/heads/main/apis.yml
apis:
- aid: glama-ai:glama-ai-gateway
  name: Glama AI Gateway
  tags:
  - AI
  - Artificial Intelligence
  - LLM Gateway
  - OpenAI Compatible
  - Multi-Provider
  humanURL: https://glama.ai/ai/gateway
  baseURL: https://gateway.glama.ai/v1
  properties:
  - url: https://glama.ai/ai/gateway
    type: Documentation
  - url: https://glama.ai/ai/models
    type: Documentation
  - url: https://gateway.glama.ai/v1
    type: BaseURL
  description: OpenAI-compatible LLM gateway exposing 90+ models from OpenAI, Anthropic, Google, DeepSeek, Mistral, xAI, Moonshot,
    Alibaba (Qwen), Cohere, and Perplexity behind a single base URL (`https://gateway.glama.ai/v1`). Drop-in replacement for
    any OpenAI SDK with prompt caching, load balancing, fallbacks, reasoning effort levels, text streaming, web search/fetch
    tools, real-time cost analytics, consolidated billing, and no rate limits under 1B tokens/day. ~40ms median overhead and
    99% uptime SLO.
- aid: glama-ai:glama-mcp-gateway
  name: Glama MCP Gateway
  tags:
  - MCP
  - Model Context Protocol
  - Gateway
  - JSON-RPC
  - OAuth
  - Observability
  humanURL: https://glama.ai/mcp/gateway
  properties:
  - url: https://glama.ai/mcp/gateway
    type: Documentation
  - url: https://glama.ai/mcp/faq
    type: Documentation
  description: Reverse-proxy MCP gateway that fronts hosted, remote, and custom MCP servers under a per-user endpoint of the
    form `https://glama.ai/endpoints/<connection-profile>/mcp`. The gateway appears as an MCP server to the client and as
    an MCP client to upstream servers. Features full JSON-RPC payload logging, per-tool access control with approval flows,
    upstream tool-definition rewriting, managed OAuth 2.1 with automatic refresh, audit log export for SIEM ingestion, session
    lifecycle tracking, and usage analytics.
- aid: glama-ai:glama-mcp-registry
  name: Glama MCP Server Registry
  tags:
  - MCP
  - Model Context Protocol
  - Marketplace
  - Discovery
  - Registry
  humanURL: https://glama.ai/mcp/servers
  properties:
  - url: https://glama.ai/mcp/servers
    type: Documentation
  - url: https://glama.ai/mcp/connectors
    type: Documentation
  - url: https://glama.ai/mcp/tools
    type: Documentation
  - url: https://glama.ai/mcp/clients
    type: Documentation
  - url: https://glama.ai/mcp/methodology
    type: Documentation
  description: Comprehensive index of 23,000+ MCP servers, 4,000+ managed connectors, and 169,000+ tools spanning 86 curated
    categories. Every server is maintainer-verified, continuously rebuilt from source, and scored for quality and safety.
    The registry is positioned as a superset of the official Model Context Protocol Registry with tool-level search, browser-based
    testing, one-click install or deploy, and quality scoring informed by Glama's own research on tool-definition quality.
- aid: glama-ai:glama-mcp-hosting
  name: Glama MCP Hosting
  tags:
  - MCP
  - Model Context Protocol
  - Hosting
  - Deployment
  humanURL: https://glama.ai/mcp/hosting
  properties:
  - url: https://glama.ai/mcp/hosting
    type: Documentation
  - url: https://glama.ai/blog/2025-09-10-glama-mcp-server-hosting
    type: Blog
  - url: https://glama.ai/blog/2025-12-06-mcp-hosting-with-persistent-storage
    type: Blog
  description: One-click managed hosting for any MCP server. Submit a GitHub repository and Glama indexes every tool, schema,
    and annotation, then offers a single click to deploy the server onto Glama's managed infrastructure with SSE security,
    persistent storage, automatic rebuilds from source, and per-tool access controls fronted by the Glama MCP Gateway. Free
    for open-source servers; paid hosting tiers (Starter/Pro/ Business) determine the number of "fast hosted MCP servers"
    included.
- aid: glama-ai:glama-mcp-inspector
  name: Glama MCP Inspector
  tags:
  - MCP
  - Model Context Protocol
  - Inspector
  - Debugging
  - Testing
  humanURL: https://glama.ai/mcp/inspector
  properties:
  - url: https://glama.ai/mcp/inspector
    type: Documentation
  - url: https://glama.ai/blog/2025-07-24-using-mcp-inspector-to-test-tools-prompts-and-resources
    type: Blog
  - url: https://glama.ai/blog/2025-07-27-how-to-use-mcp-inspector-with-remote-or-cloud-hosted-mcp-servers
    type: Blog
  description: Browser-based MCP debugger that spins up an ephemeral sandbox from a server URL. Lets developers exercise tools
    with structured inputs, view raw JSON-RPC traffic, complete OAuth flows, test prompts and resources, and share debug sessions
    via a unique URL. Works against local, remote, and Glama-hosted servers.
- aid: glama-ai:glama-chat
  name: Glama Chat and Playground
  tags:
  - AI
  - Chat
  - Playground
  - MCP
  humanURL: https://glama.ai
  properties:
  - url: https://glama.ai
    type: Documentation
  - url: https://glama.ai/ai/models
    type: Documentation
  description: All-in-one AI workspace that chats with any MCP server the user owns, routed through the Glama Gateway so every
    tool call is visible and controllable. Supports file uploads (PDFs), projects, memory, web fetch and web search tools,
    persistent storage, custom exports, and routing across 90+ models via the AI gateway. Subscription tiers include monthly
    AI credits and a number of fast hosted MCP servers.
- aid: glama-ai:glama-automations
  name: Glama Automations
  tags:
  - Automation
  - Workflows
  - AI Agents
  humanURL: https://glama.ai/blog/2025-10-22-what-are-automations
  properties:
  - url: https://glama.ai/blog/2025-10-22-what-are-automations
    type: Blog
  - url: https://glama.ai/blog/2025-09-15-automating-git-hub-pull-request-security-checks-with-glama-ai-automation-feature
    type: Blog
  description: Workflow automation primitive on the Glama platform. Composes MCP servers, the AI gateway, and Glama Chat into
    scheduled or event-triggered jobs (e.g., GitHub pull-request security checks) with full observability through the gateway
    log surface.
name: Glama
tags:
- AI
- Artificial Intelligence
- MCP
- Model Context Protocol
- LLM Gateway
- MCP Gateway
- MCP Marketplace
- AI Workspace
- Multi-Provider
- OAuth
- Observability
kind: contract
accessModel:
  pricing: unknown
  onboarding: unknown
  trial: false
  try_now: false
  public: false
  label: Unknown
  confidence: low
  source: []
  generated: '2026-07-22'
  method: derived
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/icons/glama-ai.png
access: 3rd-Party
common:
- type: DomainSecurity
  url: security/glama-ai-domain-security.yml
- type: Portal
  url: https://glama.ai
- type: Documentation
  url: https://glama.ai/mcp/faq
- type: GettingStarted
  url: https://glama.ai/mcp/faq
- type: Pricing
  url: https://glama.ai/pricing
- type: Blog
  url: https://glama.ai/blog
- type: ChangeLog
  url: https://glama.ai/release-notes
- type: StatusPage
  url: https://glama.ai/status
- type: Support
  url: https://glama.ai/support
- type: Newsletter
  url: https://glama.ai/newsletter
- type: Careers
  url: https://glama.ai/careers
- type: TermsOfService
  url: https://glama.ai/policies/terms-of-service
- type: PrivacyPolicy
  url: https://glama.ai/policies/privacy-policy
- type: SecurityPolicy
  url: https://glama.ai/policies/vulnerability-disclosure
- type: Signup
  url: https://glama.ai
- type: Forums
  name: Discord
  url: https://discord.gg/glama
- type: Forums
  name: Reddit
  url: https://www.reddit.com/r/glama
- type: GitHubOrganization
  url: https://github.com/glama-ai
- type: SDKs
  name: Lightport (open-source AI gateway)
  url: https://github.com/glama-ai/lightport
- type: Tools
  name: rjsf-validator-cfworker
  url: https://github.com/glama-ai/rjsf-validator-cfworker
- type: Documentation
  name: MCP Servers
  url: https://glama.ai/mcp/servers
- type: Documentation
  name: MCP Connectors
  url: https://glama.ai/mcp/connectors
- type: Documentation
  name: MCP Tools
  url: https://glama.ai/mcp/tools
- type: Documentation
  name: MCP Clients
  url: https://glama.ai/mcp/clients
- type: Documentation
  name: MCP Hosting
  url: https://glama.ai/mcp/hosting
- type: Documentation
  name: MCP Inspector
  url: https://glama.ai/mcp/inspector
- type: Documentation
  name: MCP Gateway
  url: https://glama.ai/mcp/gateway
- type: Documentation
  name: MCP Registry Methodology
  url: https://glama.ai/mcp/methodology
- type: Documentation
  name: AI Gateway
  url: https://glama.ai/ai/gateway
- type: Models
  url: https://glama.ai/ai/models
- type: Documentation
  name: Quickstart - Publish an MCP Server
  url: https://glama.ai/blog/2026-01-24-quickstart-publishing-a-server
- type: Documentation
  name: Official MCP Registry server.json requirements
  url: https://glama.ai/blog/2026-01-24-official-mcp-registry-serverjson-requirements
- type: Documentation
  name: Model Context Protocol
  url: https://modelcontextprotocol.io
- type: Plans
  url: https://glama.ai/pricing
  data:
  - id: starter
    name: Starter
    entries:
    - geo: US
      unit: 1
      label: User
      limit: 1
      price: 9
      metric: user
      timeFrame: month
      description: Monthly subscription including $9 of AI credits.
    elements:
    - name: Chat and API access with no rate limits
    - name: 3 fast hosted MCP servers ($4 per additional)
    - name: 100k logs per month ($9 per additional 100k)
    - name: 30-day log retention
    description: Personal use focused tier.
  - id: pro
    name: Pro
    entries:
    - geo: US
      unit: 1
      label: User
      limit: 1
      price: 26
      metric: user
      timeFrame: month
      description: Monthly subscription including $26 of AI credits.
    elements:
    - name: Everything in Starter
    - name: File uploads (PDFs), projects, memory, web fetch and web search tools
    - name: 10 fast hosted MCP servers ($3 per additional)
    - name: Persistent storage
    - name: 100k logs per month ($6 per additional 100k)
    - name: 90-day log retention
    description: Collaboration-oriented tier.
  - id: business
    name: Business
    entries:
    - geo: US
      unit: 1
      label: User
      limit: 1
      price: 80
      metric: user
      timeFrame: month
      description: Monthly subscription including $80 of AI credits.
    elements:
    - name: Everything in Pro
    - name: Custom exports and priority support
    - name: 30 fast hosted MCP servers ($2 per additional)
    - name: 100k logs per month ($3 per additional 100k)
    - name: 180-day log retention
    description: Automated-workflows tier.
  - id: open-source-hosting
    name: Open-Source MCP Hosting
    entries:
    - geo: US
      unit: 1
      label: Server
      price: Free
      metric: server
      timeFrame: month
      description: Free hosting for open-source MCP servers.
    description: Free deployment and hosting on Glama for any open-source MCP server.
  name: Plans
- type: Features
  data:
  - MCP server registry indexing 23,000+ servers, 4,000+ connectors, and 169,000+ tools across 86 categories
  - Maintainer-verified and continuously rebuilt MCP servers with quality and safety scoring
  - Browser-based MCP Inspector with ephemeral sandbox, JSON-RPC trace view, OAuth flow handling, and shareable sessions
  - One-click MCP server hosting from any GitHub repository with persistent storage and SSE security
  - Free open-source MCP server hosting
  - MCP Gateway with per-user endpoint, full JSON-RPC payload logging, per-tool access control, OAuth 2.1 credential management,
    audit log export, and session lifecycle tracking
  - Upstream tool-definition rewriting at the gateway layer
  - OpenAI-compatible AI gateway at `https://gateway.glama.ai/v1`
  - 90+ models across OpenAI, Anthropic, Google, DeepSeek, Mistral, xAI, Moonshot (Kimi), Alibaba (Qwen), Cohere, and Perplexity
  - Drop-in SDK compatibility with OpenAI SDKs and LangChain
  - Prompt caching, load balancing, automatic fallbacks, and reasoning-effort presets (low / medium / high)
  - Web search and web fetch tools across the gateway
  - Real-time cost analytics and consolidated billing with no markup on listed provider prices
  - ~40ms median gateway latency and stated 99% uptime
  - No rate limits under 1B tokens/day; higher quotas negotiable within 48 hours
  - Glama Chat playground with file uploads (PDFs), projects, memory, persistent storage, custom exports, and any MCP server
    the user owns
  - Automations primitive composing MCP servers, the AI gateway, and Chat into scheduled or triggered workflows
  - MCP Sampling support (servers can request LLM completions during execution)
  - Progress notifications, audio/image responses, and elicitations support in MCP
  - Lightport, an open-source lightweight gateway that makes LLM providers OpenAI-compatible
  - Three commercial plans (Starter $9, Pro $26, Business $80) bundling AI credits and fast hosted MCP servers
  sources:
  - https://glama.ai
  - https://glama.ai/mcp/gateway
  - https://glama.ai/mcp/servers
  - https://glama.ai/mcp/hosting
  - https://glama.ai/mcp/inspector
  - https://glama.ai/ai/gateway
  - https://glama.ai/ai/models
  - https://glama.ai/pricing
  - https://glama.ai/release-notes
  - https://glama.ai/blog
  updated: '2026-05-25'
created: '2026-05-25T00:00:00.000Z'
modified: '2026-05-25'
position: Consuming
description: Glama is an all-in-one AI workspace combining an MCP server marketplace and registry, a managed MCP gateway,
  one-click MCP server hosting, an OpenAI-compatible LLM gateway routing 90+ models across OpenAI, Anthropic, Google, DeepSeek,
  Mistral, xAI and other providers, and Glama Chat with built-in projects, memory, web tools, and automations. The Glama MCP
  Registry indexes 23,000+ MCP servers, 4,000+ managed connectors, and 169,000+ tools across 86 categories, and Glama positions
  itself as a superset of the official MCP Registry, with maintainer verification, continuous rebuilds, and capability-level
  quality scoring. The Glama MCP Gateway fronts every hosted, remote, or custom MCP server behind a per-user endpoint with
  full JSON-RPC logging, per-tool access control, managed OAuth 2.1, and audit log export. The Glama AI Gateway is OpenAI-compatible
  at `https://gateway.glama.ai/v1` and provides prompt caching, load balancing, fallbacks, reasoning effort levels, web search
  and fetch tools, real-time cost analytics, and consolidated billing with no markup.
maintainers:
- FN: Kin Lane
  email: info@apievangelist.com
  X: apievangelist
  url: https://apievangelist.com
specificationVersion: '0.16'