This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Every AI model, inside the app you already use.
Ask GPT from Claude. Ask Claude from ChatGPT. Make an image with Nano Banana from Codex.
10 free requests to try. After that, your ChatGPT plan or your own API keys pay.
Switch models, not apps. Pick the model for each task: plan in Claude, have GPT check the migration, make the cover image with Nano Banana. Every answer says what paid for it.
There are two ways to run it:
Hosted (polydev-hosted): nothing to install. The app connects to https://polydev.ai/mcp
and you sign in to Polydev in the browser the first time. Answers are paid by your own API keys,
your OpenRouter key or Polydev's 10 free requests.
Local (polydev): runs the npm client polydev-ai on your machine (Node.js 20 or newer). On top
of the hosted options it can use your ChatGPT plan (Sign in with ChatGPT), run the Claude Code or
Codex CLI you are already signed in to, and make images.
/plugin marketplace add backspacevenkat/polydev-claude-code-plugin
/plugin install polydev-hosted@polydev-marketplaceThen run /mcp, choose polydev and sign in. For the local version, install
polydev@polydev-marketplace instead; its first call opens the browser to sign in.
Without the plugin:
claude mcp add --transport http polydev https://polydev.ai/mcpRelated MCP server: multi-ai-collab
codex plugin marketplace add backspacevenkat/polydev-claude-code-plugin
codex plugin add polydev-hosted@polydevOr add the server directly:
codex mcp add polydev --url https://polydev.ai/mcp
codex mcp login polydevgemini extensions install https://github.com/backspacevenkat/polydev-claude-code-pluginThen run /mcp auth polydev.
See polydev.ai/install for one-click links and the settings for each app.
Hosted (https://polydev.ai/mcp):
Tool | What it does |
| Sends your prompt to one model or a list (for example |
| Lists the models you can reach and what pays for each |
| Searches recent posts on X (read-only; needs your own xAI key) |
| Records which answer helped most |
get_perspectives still works as the older name of ask_model.
The local client (polydev-ai 1.12.0) has the same tools plus get_auth_status, login,
chatgpt_plan_status, connect_chatgpt_plan (Continue with ChatGPT, so your ChatGPT plan pays for
OpenAI models) and generate_image.
The slash command /polydev-hosted:ask <question> (or /polydev:ask for the local version) gathers the
relevant context and calls ask_model. Name a model to ask just that one; name several to compare them.
The app calls Polydev only when you ask for it or agree to it.
Prompts you send through Polydev are passed to the model providers you chose. Don't include secrets. Disconnect an app at any time under Connected apps. See the privacy policy.
MIT, see LICENSE.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Ask GPT, Claude, Gemini and Grok at once: one consensus answer plus where they agree and split.
Every AI model, inside the app you already use: ask GPT, Claude, Gemini or Grok, and see what paid.
One thread for every AI assistant you use. Share tasks and context across ChatGPT, Claude and Grok.
Swift AI: Access a wide range of AI models⚡, including OpenAI 🤖,DeepSeek 🔍, Claude 🧠, Gemini 🌟, and.
