Google Gemini
Connect Sythoria to Google's Gemini API to chat with Google's most capable models. Gemini offers large context windows and strong performance on reasoning, code, and multilingual tasks.
Prerequisites
- A Google account
- An API key from Google AI Studio
Setup
- Go to aistudio.google.com/apikey
- Click Create API Key and copy it
- In Sythoria, open Settings from the sidebar
- Select Google Gemini from the Provider Preset dropdown — the API base URL and default model fill in automatically
- Paste your API key into the API Key field
- The connection indicator should turn green within a few seconds
Your API key is stored securely in your OS keychain. It is never sent to Sythoria's servers. See Privacy & Security.
Default configuration
| Setting | Value |
|---|
| API Base | https://generativelanguage.googleapis.com/v1beta/openai/chat/completions |
| Default Model | gemini-2.5-pro |
| API Format | OpenAI-compatible (Google's compatibility layer) |
| Streaming | Supported |
Available models
| Model ID | Best for | Context window |
|---|
gemini-1.5-pro | Most capable — complex reasoning, long documents, deep analysis | 1M |
gemini-2.0-flash | Fast and efficient — everyday tasks, quick Q&A | 1M |
gemini-1.5-flash | Previous generation flash — general-purpose, cost-effective | 1M |
How it works: OpenAI-compatible endpoint
Sythoria uses Google's OpenAI-compatible endpoint (/v1beta/openai/chat/completions). This means the experience is identical to using OpenAI's API — same request format, same streaming protocol, same response structure. No special configuration needed.
Pricing
Google offers a free tier with rate limits for development and light use. For production workloads, see ai.google.dev/pricing for pay-per-token pricing.
Troubleshooting
| Error | Cause | Fix |
|---|
| 403 Forbidden | API key lacks access to the requested model | Check your Google Cloud project settings and enable the Generative Language API |
| 400 Invalid API key | Key is malformed or revoked | Generate a new key at aistudio.google.com/apikey |
| 429 Rate Limit | Exceeded free tier quota | Wait and retry, or upgrade to a paid plan |
| Model not found | Model ID is incorrect | Use the exact model IDs from the table above |
| Connection indicator red | Network issue or invalid API base | Verify the API base URL is the OpenAI-compatible endpoint and click Refresh |
