Get API

Nano Banana 2 Lite

Gemini 3.1 Flash Lite Image Image Generation is available in the Business API catalog.

google/gemini-3.1-flash-lite-image/text-to-image

Examples

representative generations · starting points to seed the playground

Multiplication with one absence: forty-eight identical tomato-red #B8352C lip balm tubes standing in a perfect 6x8 grid on a field painted the same tomato red, photographed straight down — red-on-red world, the caps' soft speculars forming a rhythmic pattern — with exactly ONE empty slot in the grid where a tube is missing, the absence as the wit. Flat overhead soft light, milky low contrast, fine film grain, matte. No text, no logos, no HDR.

Overhead radial design: wooden matches arranged as sunburst rays around a squat red candle on a set painted deep mustard — every match pristine except one burnt curl among the rays; the radial geometry and match-head dots the graphic, soft even overhead light. Soft light, slightly muted but committed color, lifted blacks, low contrast, fine 35mm film grain, matte print finish. No text, no logos, no HDR, no harsh contrast.

Vertical full-length editorial fashion snapshot in a quiet cream apartment. Frame a single young adult woman almost symmetrically from head to shoes, standing straight against a blank ivory wall, her body centered slightly left of the vertical axis. Preserve generous pale wall around her. A narrow sunlit wall return runs down the far left; a round chrome wall sconce sits high at upper right; the curved edge of a dark walnut table enters from the lower right with a small glossy oxblood-red sculptural vase resting on two stacked books. The subject faces the camera with a calm, self-possessed expression, shoulders relaxed and both hands hidden behind her waist. Give her a distinct face, light olive skin and a blunt dark-auburn shoulder-length cut. Dress her in a fitted black mini cocktail dress with narrow straps, asymmetric nude-mesh cutouts across the torso and one ruched sheer side panel, a delicate pearl-and-metal choker, and pointed low leopard-print slingback shoes. Keep the long-legged fashion silhouette and modest styling. Shoot at eye level with a normal 50mm lens from several meters away, camera perfectly upright, full figure uncropped, minimal perspective distortion. Soft daylight enters from frame left and creates a rectangular pale glow on the wall, while the face and dress remain gently modeled. Use warm ivory, black, walnut and deep red; slightly desaturated color, low contrast, creamy highlights, open shadows, soft edges, fine 35mm grain, faint halation and a subtly aged color-negative print. No dramatic lighting, no glossy studio polish, no text or logo.

Late-1990s 35mm color-negative street-fashion snapshot with cool green shade, cyan-charcoal asphalt, restrained rusty brown and aged ivory, dense but gently lifted blacks, soft highlight rolloff, fine visible grain, slight scan noise and modest lens softness. Keep the frame tactile and naturally color-rich, not clean or glossy. Vertical 4:5 with this exact visual topology. A slender androgynous young man in his early twenties with long loose dark-chestnut hair sits sideways on a short gray steel bollard at the right-center of a leafy residential street. His torso faces slightly left; one elbow rests near his crossed knees and one knuckle touches his lips in a distant, pensive expression. His crossed legs project diagonally toward the lower-left, and the toe of his upper black slingback shoe rests precisely on the silver cap of a black fire hydrant in the lower-left foreground. He wears a loose faded cinnamon-brown overshirt, a flowing ankle-length chalky-ivory wrap skirt with one thin dark hem line, and carries a slouchy pale-gray shoulder bag against the black parked car immediately behind him. Preserve the separate fire hydrant, seating bollard, black car, blue-gray car passing on the left, dark iron fence, dense tree canopy and small transit sign behind his head. Handheld eye-level 35mm lens, close full seated figure, slightly awkward candid crop. Change the face and gender identity, but do not change the crossed-leg geometry, hand-to-mouth gesture, foreground hydrant relationship, object count, camera position or moody street-fashion energy. Soft Pika brand finish: slightly desaturated, lower digital microcontrast, natural skin and fabric texture. No HDR, commercial retouching, substituted key objects, simplified background, logo, readable text or watermark.

Per-model pricing

Transparent, usage-based rates for every Google model.

Gemini Omni Flash

Video

Text to Video

Video output$17.50 / 1M tok
Input$1.50 / 1M tok
Output$9 / 1M tok

Image to Video

Video output$17.50 / 1M tok
Input$1.50 / 1M tok
Output$9 / 1M tok

Reference to Video

Video output$17.50 / 1M tok
Input$1.50 / 1M tok
Output$9 / 1M tok

Video to Video

Video output$17.50 / 1M tok
Input$1.50 / 1M tok
Output$9 / 1M tok

Veo 3.1

Video

Text to Video

720pno audio$0.20 / sec
1080pno audio$0.20 / sec
720pwith audio$0.40 / sec
1080pwith audio$0.40 / sec
4kno audio$0.40 / sec
4kwith audio$0.60 / sec

Video Extension

no audio$0.20 / sec
with audio$0.40 / sec

Veo 3.1 Fast

Video

Text to Video

720pno audio$0.08 / sec
720pwith audio$0.10 / sec
1080pno audio$0.10 / sec
1080pwith audio$0.12 / sec
4kno audio$0.25 / sec
4kwith audio$0.30 / sec

Video Extension

no audio$0.08 / sec
with audio$0.10 / sec

Veo 3.1 Lite

Video

Image to Video

720pno audio$0.03 / sec
720pwith audio$0.05 / sec
1080pno audio$0.05 / sec
1080pwith audio$0.08 / sec

Nano Banana Pro

Image

Image to Image

Image output$120 / 1M tok
Input$2 / 1M tok
Output$12 / 1M tok

Text to Image

Image output$120 / 1M tok
Input$2 / 1M tok
Output$12 / 1M tok

Nano Banana 2

Image

Image to Image

Image output$60 / 1M tok
Input$0.50 / 1M tok
Output$3 / 1M tok

Text to Image

Image output$60 / 1M tok
Input$0.50 / 1M tok
Output$3 / 1M tok

Nano Banana 2 Lite

Image

Image to Image

Image output$30 / 1M tok
Input$0.25 / 1M tok
Output$1.50 / 1M tok

Text to ImageCurrent

Image output$30 / 1M tok
Input$0.25 / 1M tok
Output$1.50 / 1M tok

Lyria 3 Pro

Audio • Text to Audio

$0.064 / request

Lyria 3

Audio • Text to Audio

$0.032 / request

Lyria 2

Audio • Music

$0.048 / request

Gemini 3.6 Flash

LLM

Input$1.20 / 1M tok
Output$6 / 1M tok
Cached input$0.12 / 1M tok

Gemini 3.5 Flash-Lite

LLM

Input$0.24 / 1M tok
Output$2 / 1M tok
Cached input$0.024 / 1M tok

Gemini 3.1 Pro

LLM

Input$1.60 / 1M tok
Output$9.60 / 1M tok
Cached input$0.16 / 1M tok