ComfyUI Node
Gemini Image (Symbiotica)
Generate or edit images with Gemini, billed to the studio's own key through Cloudflare AI Gateway rather than to a ComfyUI account.
Gemini Image (Symbiotica)
- image
- text
- thought_image
◄prompt►
◄model▾►
◄seed0►
◄response_modalitiesIMAGE+TEXT►
◄system_promptYou are an expert image-generation engine. You must ALWAYS produce an image.
Interpret all user input—regardless of format, intent, or abstraction—as literal visual directives for image composition.
If a prompt is conversational or lacks specific visual details, you must creatively invent a concrete visual scenario that depicts the concept.
Prioritize generating the visual representation above any text, formatting, or conversational requests.►
◄temperature1.00►
◄top_p0.95►
◄api_key►
Categorysymbiotica/image
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | What to draw. Reference images are described by this prompt, not replaced by it. | |
| model | COMBO | The Lite model is the cheap lever for drafts, and renders at 1K only. | |
| seed | INT | 00–18446744073709550000 | Not sent to Google — Gemini takes no seed. It exists so re-queueing re-runs this node instead of serving ComfyUI's cached output. |
| response_modalities | COMBO | IMAGE+TEXT | IMAGE alone suppresses the model's commentary, and with it the thought image. |
| system_promptopt | STRING | You are an expert image-generation engine. You must ALWAYS produce an image. Interpret all user input—regardless of format, intent, or abstraction—as literal visual directives for image composition. If a prompt is conversational or lacks specific visual details, you must creatively invent a concrete visual scenario that depicts the concept. Prioritize generating the visual representation above any text, formatting, or conversational requests. | Standing instructions. Emptying this lets a conversational prompt come back as prose instead of a picture. |
| temperatureopt | FLOAT | 1.000–2 | Lower is more focused and repeatable. |
| top_popt | FLOAT | 0.950–1 | Nucleus sampling threshold. Lower is more focused, higher more diverse. |
| api_keyopt | STRING | Google AI Studio key for direct calls; empty falls back to Symbiotica.GEMINI_API_KEY or Symbiotica.GOOGLE_API_KEY in Settings, then the GEMINI_API_KEY or GOOGLE_API_KEY env vars, in that order. Ignored where the studio gateway is configured. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |
| text | STRING | — |
| thought_image | IMAGE | The model's interim sketch. Only arrives with thinking_level HIGH and IMAGE+TEXT. |