gemini

Call Gemini 2.5 and 3 series models through RunAPI using Gemini contents clients. Use for Gemini chat, streaming, multimodal input, grounding, structured output, reasoning, or an e…

RunAPI

@runapi-ai

What This Skill Does

Provides access to Google Gemini 2.5 and 3 series models through RunAPI, supporting both OpenAI-compatible and native Gemini contents request styles for chat, streaming, multimodal vision, and structured output. Requires a RunAPI API key and works with existing OpenAI or Gemini SDKs by pointing them at the RunAPI base URL.

Replaces the need for a direct Google AI API key and separate Gemini SDK setup by letting you use Gemini models through a single RunAPI token with either OpenAI or Gemini client libraries.

When to Use It

  • Send chat completion requests to Gemini 2.5 or 3 models using the OpenAI Python or TypeScript SDK
  • Stream long-form text generation from Gemini models without blocking server threads
  • Analyze images with multimodal vision input using Gemini's contents request format
  • Generate structured JSON output from Gemini with controlled reasoning effort
  • Point an existing Gemini SDK application at RunAPI as the base URL for unified billing and logging
  • Use Google Search grounding with Gemini responses for factually grounded answers

Install

$ openclaw skills install @runapi-ai/runapi-gemini

Gemini on RunAPI

Use Gemini contents at https://runapi.ai with generateContent or streamGenerateContent as the primary protocol.

Primary protocol recipe

Authenticate

Set GEMINI_API_KEY to a RunAPI API key and the Gemini client base URL to https://runapi.ai. Raw requests send the key with x-goog-api-key.

Send request

curl -X POST \
  "https://runapi.ai/v1beta/models/gemini-3.7-flash:streamGenerateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"contents":[{"role":"user","parts":[{"text":"Summarize this review."}]}]}'

Use generateContent only when the selected model supports synchronous output; gemini-3.7-flash accepts only streamGenerateContent. For this model, native tools are limited to functionDeclarations and the googleSearch tool with an empty object payload; native thinking is limited to generationConfig.thinkingConfig.includeThoughts and thinkingLevel of low or high.

Verify result

Require candidate content, terminal finishReason, and usageMetadata. A stream is complete only after the terminal candidate and Usage; an HTTP 2xx without both is incomplete.

Stop boundaries

Correct a rejected shape once using the structured error. Retry transport once only before any candidate or Usage and when replay is safe. Record a terminal error and stop without changing model or protocol.

Compatibility protocols

Load compatibility protocols only when an existing client requires OpenAI-compatible Chat Completions.

Supported models

Model IDUse when
gemini-3.7-flashFast streaming chat; use native tools/search grounding only with Gemini contents
gemini-3.6-flashFast current-generation multimodal requests
gemini-3.5-flashFast multimodal and function-calling requests
gemini-3.1-pro-previewComplex multimodal and reasoning requests
gemini-3-flash-previewPreview flash workloads
gemini-2.5-proStable high-quality multimodal requests
gemini-2.5-flashStable fast multimodal requests

References

Top skills in this category