Unverified50% confidenceFactExact time
The native Google Gemini API uses contents instead of messages, uses role: model instead of role: assistant, and nests text inside a parts array.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGoogle's Gemini uses the SentencePiece framework combined with the Unigram language model algorithm for tokenization.77% similarUnverifiedGoogle Gemini's request format uses URL query parameters to pass the API key, and the request body uses a nested structure of contents→parts→text.74% similarUnverifiedThis workflow calls Google Gemini to generate the body copy, hashtags, source notes, and accompanying image for a LinkedIn post.73% similarUnverifiedGoogle has long deployed a Function Calling interface in Gemini, laying the underlying foundation for agent applications such as Spark.72% similarUnverifiedGoogle officially launched the general-purpose work agent Gemini Agent, embedded in the office suite, able to share context across email, documents, and spreadsheets, and supporting multi-agent collaboration.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1052066API
curl https://kongchang.com/api/v1/knowledge/claims/1052066MCP
get_claim(id=1052066)