[KongchangAI]
ModelGemini 3.8 Flash

Gemini-3.8-flash

A lightweight multimodal large model in the Gemini series released by Google DeepMind, supporting audio and text understanding, and focused on efficient inference and low-cost deployment.

Timeline (last 90 days)

Nov 10

Preliminary evaluation on Gemini-3.8-flash shows: when no auxiliary information is provided, accuracy is 29.3%.

Unverified50%
Nov 10

On Gemini-3.8-flash, accuracy rises to 31.3% when the rule is provided, and reaches 98.5% when the acoustic observation text is also provided.

Unverified50%

All Facts (2)

Source Articles