7 Best Runway MCP Alternatives in 2026 (Free & Paid)
Runway MCP is hosted Runway server to generate images and video with your Runway plan credits, priced at free plan, paid from $15/mo. If it is too expensive, missing a feature or simply not for you, these are the strongest alternatives we have reviewed.
1
Hosted fal.ai server to search, price and run 1,000+ image, video and audio models
8.2/10
What it is The fal MCP server is fal.ai's official hosted endpoint at mcp.fal.ai/mcp. It gives an AI assistant access to fal's catalogue of more than 1,000 generative models (image, video, audio, 3D), plus fal's documentation, so the agent can find a model, read its input schema, check the price and run it. Key features - search_models and recommend_model to… Read more →
Pros
- 1,000+ models behind one key
- Price check tool before you generate
- Hosted, with OAuth for Cursor, Windsurf and ChatGPT
- Async job tools suit long video renders
Cons
- Pay-per-output only, no free tier listed
- Licences differ per model, so check commercial terms each time
- Easy to run up costs with video models
Pricing: Freemium · image-generation video-generation model-marketplace official hosted
2
ElevenLabs' hosted MCP server for speech generation and voice-agent management
8.0/10
What it is The ElevenLabs MCP server is the official, ElevenLabs-hosted Model Context Protocol endpoint at api.elevenlabs.io/v1/mcp. It lets an MCP client such as Claude create speech audio from text and manage ElevenLabs voice agents from a chat. It replaced the older local elevenlabs-mcp Python server, which ElevenLabs archived in August 2026 and marked deprecated in favour of the hosted… Read more →
Pros
- Official and hosted, nothing to install
- OAuth sign-in, no API key in the client config
- Regional endpoints for data residency
- Admins can restrict which tools users may call
Cons
- Hosted tool list is focused on voice agents, narrower than the old local server
- Local open-source server is archived and deprecated
- Free plan output is non-commercial
Pricing: Free plan, paid from $6/mo · text-to-speech voice voice-agents official hosted
3
Comfy Org's official MCP for local ComfyUI and Comfy Cloud workflows
★ 234 · updated 2026-09-20
7.9/10
What it is Comfy MCP is the first-party Model Context Protocol integration from Comfy Org, the company behind ComfyUI, launched in June 2026. It comes in two forms: comfy-mcp, an open-source local server that drives your own ComfyUI install, and Comfy Cloud MCP at cloud.comfy.org/mcp, which runs workflows on Comfy's GPUs with no local install. Key features - Search templates,… Read more →
Pros
- Official from Comfy Org
- Local and cloud options
- Can author and edit workflows, not just run them
- Spending gate before paid partner models
Cons
- Local server is beta
- Needs a running ComfyUI install locally
- Cloud GPU credits add up on video
Pricing: Open source, paid cloud from $16/mo · comfyui image-generation workflows official local
4
Hosted HeyGen server for avatar videos, video translation and templates from any agent
7.8/10
What it is HeyGen MCP is HeyGen's official remote Model Context Protocol server at mcp.heygen.com/mcp/v1/. It lets an AI agent create AI-avatar videos, translate videos and work with templates using the avatars, voices and brand kits in your HeyGen account. There is no local repository; it runs on HeyGen's infrastructure. Key features - Video Agent sessions: describe a video and… Read more →
Pros
- No API key, OAuth only
- Uses existing plan credits, no separate API billing
- Video translation and templates exposed
- Custom avatars and brand kits available
Cons
- Free plan output is watermarked and limited to 3 short videos
- Credits burn quickly on long videos
- Custom agents need domain approval
Pricing: Free plan, paid from $29/mo · avatar-video video-translation official hosted oauth
5
Official Replicate server to search and run thousands of open models from a chat
7.8/10
What it is Replicate MCP is Replicate's official Model Context Protocol server. It exposes Replicate's HTTP API, so an assistant can search models, run predictions and read model metadata. It is offered as a hosted server at mcp.replicate.com with web sign-in, and as the replicate-mcp npm package you run locally. Key features - Search Replicate's catalogue by task (for example,… Read more →
Pros
- Thousands of open models
- Official, with hosted and local options
- Full HTTP API coverage
- Works with your own deployed models
Cons
- Stable npm release is old; docs push the alpha track
- Code mode is experimental and needs Deno
- Usage-based billing, no free allowance stated
Pricing: Freemium · image-generation open-models video-generation official api
6
Official MiniMax server for TTS, voice cloning, Hailuo video and image generation
★ 1.6k · MIT · updated 2026-08-20
7.4/10
What it is MiniMax MCP is the official Model Context Protocol server from MiniMax, the Chinese AI lab behind the Hailuo video models and Speech voices. It is an open-source Python server that runs locally and calls the MiniMax API, so an MCP client such as Claude Desktop, Cursor or Windsurf can produce speech, cloned voices, video clips and images… Read more →
Pros
- Official server from the model maker
- Voice, video and image in one server
- Voice cloning and voice design tools
- MIT-licensed and self-hostable
Cons
- API key must match the region host or calls fail
- Local install with uv, no hosted endpoint documented
- Video jobs are asynchronous and need polling
Pricing: Free plan, paid from $5/mo · text-to-speech video-generation hailuo voice-cloning official
7
Google Cloud sample servers for Veo, Gemini Image, Gemini TTS, Lyria and AV compositing
★ 1.2k · Apache-2.0 · updated 2026-09-25
6.9/10
What it is MCP Servers for Google Cloud Genmedia is a set of Go-based MCP servers published in Google Cloud's Creative Studio repository. Each server wraps a Vertex AI generative media API, so an agent can make video with Veo, images with Gemini Image models, speech with Gemini TTS and Chirp 3 HD, and music with Lyria, then combine them… Read more →
Pros
- Veo, Gemini Image, TTS and Lyria in one set
- AVTool for compositing
- Apache-2.0, prebuilt binaries
Cons
- Marked as not an officially supported Google product
- Requires a Google Cloud project and billing
- Docs target Gemini CLI and ADK more than other clients
Pricing: Freemium · veo lyria google-cloud video-generation text-to-speech