Nodes/WorkflowX-Configurator/NanoBanana Full API
ComfyUI Node

NanoBanana Full API

Generate or edit images with Gemini 3.1 Flash Image or Gemini 3 Pro Image at 1K, 2K, or 4K resolution.

By haroonaslam·Created 3 months ago·Updated a day ago· 10
NanoBanana Full API
  • mask
  • image_1
  • image_2
  • image_3
  • image_4
  • image_5
  • image_batch
  • text_output
api_key
model_versiongemini-3.1-flash-image
promptA majestic golden retriever, watercolor style
system_promptYou are a helpful image generation assistant.
aspect_ratio1:1
seed0
temperature1.0
top_p0.95
candidate_count1
safety_harassmentBLOCK_DEFAULT
safety_hate_speechBLOCK_DEFAULT
safety_sexualBLOCK_DEFAULT
safety_dangerousBLOCK_DEFAULT
edit_mode_enabledno
resolution4K
timeout_seconds120
show_thoughtstrue
thinking_levelhigh
CategoryWorkflowX/API

Inputs (24)

NameTypeDefaultDescription
api_keySTRINGGoogle Gemini API key. When blank, GEMINI_API_KEY then GOOGLE_API_KEY are checked.
model_versionCOMBOgemini-3.1-flash-image2 options: gemini-3.1-flash-image, gemini-3-pro-image
promptSTRINGA majestic golden retriever, watercolor style
system_promptSTRINGYou are a helpful image generation assistant.
aspect_ratioCOMBO1:110 options: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, +4
seedINT00–2147483647
temperatureFLOAT1.00–2
top_pFLOAT0.950–1
candidate_countINT11–10
safety_harassmentCOMBOBLOCK_DEFAULT5 options: BLOCK_DEFAULT, BLOCK_NONE, BLOCK_LOW_AND_ABOVE, BLOCK_MEDIUM_AND_ABOVE, BLOCK_ONLY_HIGH
safety_hate_speechCOMBOBLOCK_DEFAULT5 options: BLOCK_DEFAULT, BLOCK_NONE, BLOCK_LOW_AND_ABOVE, BLOCK_MEDIUM_AND_ABOVE, BLOCK_ONLY_HIGH
safety_sexualCOMBOBLOCK_DEFAULT5 options: BLOCK_DEFAULT, BLOCK_NONE, BLOCK_LOW_AND_ABOVE, BLOCK_MEDIUM_AND_ABOVE, BLOCK_ONLY_HIGH
safety_dangerousCOMBOBLOCK_DEFAULT5 options: BLOCK_DEFAULT, BLOCK_NONE, BLOCK_LOW_AND_ABOVE, BLOCK_MEDIUM_AND_ABOVE, BLOCK_ONLY_HIGH
edit_mode_enabledCOMBOno2 options: no, yes
resolutionCOMBO4K3 options: 1K, 2K, 4K
timeout_secondsINT1201–3600Maximum time to wait for the Google API response.
show_thoughtsBOOLEANtrueRequest and include model thought text in text_output.
thinking_levelCOMBOhighGemini 3.1 Flash Image only. Gemini 3 Pro Image uses its model-managed thinking level.
maskoptMASK
image_1optIMAGE
image_2optIMAGE
image_3optIMAGE
image_4optIMAGE
image_5optIMAGE

Outputs (2)

NameTypeDescription
image_batchIMAGE
text_outputSTRING