Connect OmniDino Gemini models in Cherry Studio.
Use Cherry Studio’s Gemini Provider to access Gemini text and multimodal models through OmniDino’s Gemini-compatible endpoint, all from one desktop client.
Manage multiple model services from one client.
Cherry Studio supports OpenAI, Anthropic, Gemini-compatible providers, custom models, knowledge bases, MCP, translation, and multi-model conversations.
Bring models, API Keys, and conversation workflows into one desktop app.
OmniDino handles model routing, balance, and usage billing. Cherry Studio handles desktop conversations, model selection, knowledge bases, and MCP. Create a dedicated API Key for Cherry Studio and set a sensible spending limit.
Install Cherry Studio from the official GitHub repository.
Cherry Studio Desktop
Download the installer that matches your operating system and CPU architecture from GitHub Releases.
OmniDino API Key
Sign in to the OmniDino console, create a dedicated API Key, and confirm that the Key has balance, model permissions, and the correct group.
From Provider settings to the first successful response.
Open the Gemini Provider
Open Cherry Studio model-service settings, search for Gemini, and select the matching Provider.
Enter the OmniDino API Key
Use the dedicated Key created in the OmniDino console.
Enter the API URL
Set the API URL to https://api.omnidino.com/v1.
Fetch or Add Models
Fetch the model list, or manually add the exact Gemini model IDs available in OmniDino.
Select and Verify
Return to the chat page, select the Gemini model, run a minimal text test, and then test image input.
Recommended OmniDino configuration.
Field names may vary slightly between Cherry Studio versions, but Provider, API Key, API URL, and model list are the core settings.
Complete These Fields
The Gemini Provider is for models exposed by OmniDino through a Gemini-compatible protocol.
https://api.omnidino.com/v1. If your channel requires native Gemini paths, follow the protocol configured in OmniDino and verify it with an actual request.Provider: Gemini
API Key: YOUR_OMNIDINO_API_KEY
API Base: https://api.omnidino.com/v1
Models: Fetch Models or add actual Gemini model IDs
Status: EnabledTest text first, then move to advanced features.
Verification Order
Reply with only:
OmniDino Gemini connection successful.
Do not call tools or read files.Which tasks are best suited to the Gemini Provider?
Multimodal Understanding
Use a model with image input to analyze screenshots, images, and visual material.
Long Documents
Ideal for research material, reports, manuals, and large-context tasks.
Coding Tasks
Explain code, generate approaches, and assist development, depending on the selected model.
Fast Models
Flash-class models are often suitable for frequent, low-latency, cost-sensitive tasks.
Complex Reasoning
Pro-class models are designed for harder tasks but may be slower and more expensive.
Image Generation Models
Image output differs from chat and image understanding. Use a supported model and matching client feature.
Troubleshoot connection problems in order.
https://api.omnidino.com/v1 and that the Key is valid and funded. If automatic discovery fails, add the model ID manually.Unlock Gemini’s multimodal capabilities in Cherry Studio.
Verify the text route first, then test image input, long context, and advanced tools to avoid unnecessary protocol problems.