Expired50% confidenceFactExact time
The training of the Gemini series of models is based on JAX.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/12/2026
First Seen
Valid until: 10/10/2026(expired)
Sources
AI全栈方法解读：谷歌从芯片到应用的垂直整合战略
rss6/29/2026
Related Claims
UnverifiedThe Gemini Interactions API was released after Gemini's last training run, causing the model to frequently generate code using the older Gemini 2.0.78% similarUnverifiedGemini mixes text, image, audio, and video data as input from the pre-training stage to achieve native multimodal training.74% similarVerifiedGemini adopts a native multimodal architecture, jointly training the Transformer with mixed multimodal data from the pre-training stage, with tokens of different modalities processed together in the same sequence.73% similarVerifiedThe Gemini series jointly models multiple modalities from the training stage, rather than appending a visual encoder as solutions like GPT-4V do.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/491106API
curl https://kongchang.com/api/v1/knowledge/claims/491106MCP
get_claim(id=491106)