HopBase DocsGemini chat
Call Gemini models through the OpenAI compatible protocol and Interactions API.
Gemini chat models use the OpenAI Chat Completions compatible protocol: POST https://api.hop-base.com/v1/chat/completions with Authorization: Bearer sk-your-key. Existing OpenAI SDKs and third-party clients only need a Gemini model ID.
Available models
| Family | Model ID | Use case |
|---|---|---|
| Gemini Omni | gemini-omni-flash-preview | Text, image, and video input; video output |
| Gemini 3.6 | gemini-3.6-flash | Latest Flash model; currently under observation |
| Gemini 3.5 | gemini-3.5-flash / gemini-3.5-flash-lite | Recommended default / low cost |
| Gemini 3.1 Pro | gemini-3.1-pro-preview / gemini-3.1-pro-preview-customtools | Advanced reasoning preview / tool-use preview |
| Gemini 3.1 | gemini-3.1-flash-lite / gemini-3.1-flash-lite-preview | Lightweight / lightweight preview |
| Gemini 3 | gemini-3-flash-preview | Gemini 3 Flash preview |
| Gemini 2.5 | gemini-2.5-pro / gemini-2.5-flash / gemini-2.5-flash-lite | Stable models |
Use gemini-3.5-flash by default. Gemini 2.5 stable models are the mature fallback. gemini-3.6-flash is newly available and remains under observation. Treat every model ID containing preview as a preview without a stability guarantee.
Rates vary by plan group. The signed-in model catalog for the current key is authoritative.
curl example
curl https://api.hop-base.com/v1/chat/completions \
-H "Authorization: Bearer sk-your-key" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.5-flash",
"messages": [{"role": "user", "content": "Hello"}]
}'Streaming with SSE (add "stream": true)
curl https://api.hop-base.com/v1/chat/completions \
-H "Authorization: Bearer sk-your-key" \
-H "Content-Type: application/json" \
-d '{"model": "gemini-3.1-pro-preview", "stream": true,
"messages": [{"role": "user", "content": "Hello"}]}'Gemini-native payload passthrough
Existing Gemini SDK request bodies can keep their native contents field. The same endpoint detects the format and forwards it as a native Gemini request.
curl https://api.hop-base.com/v1/chat/completions \
-H "Authorization: Bearer sk-your-key" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.5-flash",
"contents": [{"role": "user", "parts": [{"text": "Hello"}]}]
}'Gemini Omni video generation (native Interactions)
gemini-omni-flash-preview accepts text, images, and video, and can output video directly. Text-to-video only needs model and input. Use POST https://api.hop-base.com/v1beta/interactions for the complete native video response. The Chat Completions endpoint remains available for text chat but should not be used to receive the complete video payload. POST /v1/interactions remains as a compatibility alias.
curl https://api.hop-base.com/v1beta/interactions \
-H "Authorization: Bearer sk-your-key" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-omni-flash-preview",
"input": "Generate a 3-second video of a red ball slowly rolling on a white background"
}'The video is in steps[].content[], with data holding a base64-encoded MP4. usage.output_tokens_by_modality breaks the output tokens down by modality for your information — those numbers are already inside total_output_tokens, which is what billing uses. See Gemini Omni video for the full contract and pricing.
{
"status": "completed",
"model": "gemini-omni-flash-preview",
"steps": [{
"type": "model_output",
"content": [{
"type": "video",
"mime_type": "video/mp4",
"data": "<base64-mp4>"
}]
}],
"usage": {
"total_input_tokens": 20,
"total_output_tokens": 17819,
"output_tokens_by_modality": [{"modality": "video", "tokens": 17376}]
}
}Input video is limited to 10 seconds. Output is 3-10 seconds at 720p and 24 FPS. State the desired duration in input. The model context is 1,048,576 tokens; preview capabilities and availability can change.
Available Gemini models vary by plan group. Call GET /v1/models with the current key and only use complete IDs returned by that response.