Replicate for Claude, Cursor, ChatGPT & agents

Replicate — run any model from the Replicate model marketplace (text-to-image, image-restoration, audio-to-text, video, language)

curl -fsSL https://syncorelabs.ai/install.sh | sh

What it does

Replicate — run any model from the Replicate model marketplace (text-to-image, image-restoration, audio-to-text, video, language). Run synchronously (waits up to 4 minutes) or async (kick off + poll), cancel running jobs, list recent predictions, browse and search the public model catalog, list curated collections, fetch model metadata + input schema. BYOK only — bills your Replicate account directly. Use when the user mentions Replicate, run a model, FLUX, SDXL, video model, image generation, or wants to invoke open-source ML models.

The Replicate skill is one of 75+ pre-wired Syncore integrations that any MCP-compatible client — Claude Desktop, Claude Code, Cursor, Windsurf, Codex CLI, and ChatGPT desktop — can call out of the box. You don't configure MCP servers or copy API keys into a JSON file; the Syncore daemon discovers the skill at startup and exposes its tools to whichever AI client is running.

Sample prompts

Once Syncore is installed and Replicate is connected, ask your agent something like:

• “Run Prediction prediction from Replicate.”
• “Create prediction from Replicate.”
• “Get prediction from Replicate.”

Tools available (10)

• replicate__run_prediction

  Synchronously run a model and wait for the result (up to 4 minutes). Use for fast models. For long-running jobs use create_prediction + get_prediction.

• replicate__create_prediction

  Kick off a prediction asynchronously. Returns the prediction id; poll get_prediction for status / output.

• replicate__get_prediction

  Get the current status + output (if ready) of a prediction.

• replicate__cancel_prediction

  Cancel a running prediction. Idempotent.

• replicate__list_predictions

  List your recent predictions with status + output urls.

• replicate__list_models

  List public models, optionally filtered by collection (e.g. text-to-image, image-restoration, audio-to-text).

• replicate__get_model

  Fetch one model's metadata: latest version, owner, description, input schema.

• replicate__search_models

  Free-text search the model catalog (Replicate's /models?search endpoint).

• replicate__list_collections

  List Replicate's curated model collections (text-to-image, music, etc.) for browsing.

• replicate__get_account

  Show whoami — useful for verifying the API token works without firing a prediction.

How to connect

Replicate needs a provider API key to talk to replicate. Connect it from the Syncore dashboard:

1. Run curl -fsSL https://syncorelabs.ai/install.sh | sh on macOS / Linux (or the PowerShell command on Windows).
2. Run syncore login to create your account and a device-bound encryption key.
3. Visit syncorelabs.ai/connect and authorize Replicate.
4. Run syncore update to sync the skill + credentials to every installed AI client.

Pricing

Included on every Syncore tier with a per-tier shared quota — Free covers light use; Pro at $29/mo and Ultra at $99/mo unlock larger monthly allowances. Bring your own provider API key any time to bypass the Syncore quota.

See the full Syncore pricing breakdown — Free $0/mo, Pro $29/mo, Ultra $99/mo — on the homepage.