i use the follow config. whatever resolution i input,it only generate 1k image.
I found that it’s affected by the TEXT property of responseModalities and the include_thoughts property of thinking_config. When I set responseModalities: ["IMAGE"], the image is generated correctly; however, when I set responseModalities: ["IMAGE", "TEXT"], it only generates a 1K image.
payload = {
"contents": \[{
"role": "user",
"parts": \[{"text": "Generate an image of a cat."}\]
}\],
"generationConfig": {
"responseModalities": \["IMAGE", "TEXT"\],
"imageConfig": {
"imageSize": "2K",
"aspectRatio": "1:1"
},
"thinking_config": {
"thinking_level":"high",
"include_thoughts":True
}
}
}