Gemini może generować i przetwarzać obrazy w formie rozmowy. Możesz przekazywać Gemini tekst, obrazy lub kombinację obu tych elementów, co pozwala tworzyć, edytować i ulepszać treści wizualne z niespotykaną dotąd kontrolą:
- Text-to-Image: generuj wysokiej jakości obrazy na podstawie prostych lub złożonych opisów tekstowych.
- Obraz + tekst na obraz (edycja): podaj obraz i użyj promptów tekstowych, aby dodać, usunąć lub zmodyfikować elementy, zmienić styl lub dostosować korekcję kolorów.
- Wiele obrazów do obrazu (kompozycja i przenoszenie stylu): użyj kilku obrazów wejściowych, aby skomponować nową scenę lub przenieść styl z jednego obrazu na drugi.
- Iteracyjne dopracowywanie: prowadź rozmowę, aby stopniowo dopracowywać obraz w wielu etapach, wprowadzając drobne zmiany, aż będzie idealny.
- Renderowanie tekstu w wysokiej jakości: dokładne generowanie obrazów zawierających czytelny i dobrze umieszczony tekst, idealne do logo, diagramów i plakatów.
Wszystkie wygenerowane obrazy zawierają znak wodny SynthID.
W tym przewodniku opisujemy zarówno szybki model Gemini 2.5 Flash, jak i zaawansowany model Gemini 3 Pro Preview. Znajdziesz tu przykłady możliwości od podstawowej zamiany tekstu na obraz po złożone, wieloetapowe ulepszenia, wyjście w rozdzielczości 4K i generowanie na podstawie wyszukiwania.
Wybór modelu
Wybierz model, który najlepiej pasuje do Twojego konkretnego zastosowania.
Gemini 3 Pro Image Preview (wersja testowa Nano Banana Pro) jest przeznaczony do profesjonalnego tworzenia komponentów i wykonywania złożonych instrukcji. Ten model wykorzystuje informacje z wyszukiwarki Google, domyślny proces „myślenia”, który udoskonala kompozycję przed wygenerowaniem obrazu, i może generować obrazy w rozdzielczości do 4K.
Gemini 2.5 Flash Image (Nano Banana) został zaprojektowany z myślą o szybkości i wydajności. Ten model jest zoptymalizowany pod kątem zadań o dużej liczbie zapytań i małym opóźnieniu. Generuje obrazy w rozdzielczości 1024 pikseli.
Generowanie obrazów (tekst na obraz)
Poniższy kod pokazuje, jak wygenerować obraz na podstawie opisowego prompta.
Python
from google import genai
from google.genai import types
from PIL import Image
client = genai.Client()
prompt = (
"Create a picture of a nano banana dish in a fancy restaurant with a Gemini theme"
)
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[prompt],
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("generated_image.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
"Create a picture of a nano banana dish in a fancy restaurant with a Gemini theme";
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("gemini-native-image.png", buffer);
console.log("Image saved as gemini-native-image.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("Create a picture of a nano banana dish in a " +
" fancy restaurant with a Gemini theme"),
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "gemini_generated_image.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
Java
import com.google.genai.Client;
import com.google.genai.types.GenerateContentConfig;
import com.google.genai.types.GenerateContentResponse;
import com.google.genai.types.Part;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Paths;
public class TextToImage {
public static void main(String[] args) throws IOException {
try (Client client = new Client()) {
GenerateContentConfig config = GenerateContentConfig.builder()
.responseModalities("TEXT", "IMAGE")
.build();
GenerateContentResponse response = client.models.generateContent(
"gemini-3-pro-image-preview",
"Create a picture of a nano banana dish in a fancy restaurant with a Gemini theme",
config);
for (Part part : response.parts()) {
if (part.text().isPresent()) {
System.out.println(part.text().get());
} else if (part.inlineData().isPresent()) {
var blob = part.inlineData().get();
if (blob.data().isPresent()) {
Files.write(Paths.get("_01_generated_image.png"), blob.data().get());
}
}
}
}
}
}
REST
curl -s -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{
"parts": [
{"text": "Create a picture of a nano banana dish in a fancy restaurant with a Gemini theme"}
]
}]
}' \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > gemini-native-image.png
Edytowanie obrazów (tekst i obraz na obraz)
Przypomnienie: upewnij się, że masz wymagane prawa do wszystkich przesyłanych obrazów. Nie twórz treści naruszających prawa innych osób, w tym filmów ani obrazów, które mogą zostać wykorzystane do oszustwa, nękania lub wyrządzania krzywdy. Korzystanie z tej usługi generatywnej AI podlega naszym Zasadom dotyczącym niedozwolonych zastosowań.
Poniższy przykład pokazuje przesyłanie obrazów zakodowanych w formacie base64. Więcej informacji o wielu obrazach, większych ładunkach i obsługiwanych typach MIME znajdziesz na stronie Rozumienie obrazów.
Python
from google import genai
from google.genai import types
from PIL import Image
client = genai.Client()
prompt = (
"Create a picture of my cat eating a nano-banana in a "
"fancy restaurant under the Gemini constellation",
)
image = Image.open("/path/to/cat_image.png")
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[prompt, image],
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("generated_image.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const imagePath = "path/to/cat_image.png";
const imageData = fs.readFileSync(imagePath);
const base64Image = imageData.toString("base64");
const prompt = [
{ text: "Create a picture of my cat eating a nano-banana in a" +
"fancy restaurant under the Gemini constellation" },
{
inlineData: {
mimeType: "image/png",
data: base64Image,
},
},
];
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("gemini-native-image.png", buffer);
console.log("Image saved as gemini-native-image.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
imagePath := "/path/to/cat_image.png"
imgData, _ := os.ReadFile(imagePath)
parts := []*genai.Part{
genai.NewPartFromText("Create a picture of my cat eating a nano-banana in a fancy restaurant under the Gemini constellation"),
&genai.Part{
InlineData: &genai.Blob{
MIMEType: "image/png",
Data: imgData,
},
},
}
contents := []*genai.Content{
genai.NewContentFromParts(parts, genai.RoleUser),
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
contents,
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "gemini_generated_image.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
Java
import com.google.genai.Client;
import com.google.genai.types.Content;
import com.google.genai.types.GenerateContentConfig;
import com.google.genai.types.GenerateContentResponse;
import com.google.genai.types.Part;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;
public class TextAndImageToImage {
public static void main(String[] args) throws IOException {
try (Client client = new Client()) {
GenerateContentConfig config = GenerateContentConfig.builder()
.responseModalities("TEXT", "IMAGE")
.build();
GenerateContentResponse response = client.models.generateContent(
"gemini-3-pro-image-preview",
Content.fromParts(
Part.fromText("""
Create a picture of my cat eating a nano-banana in
a fancy restaurant under the Gemini constellation
"""),
Part.fromBytes(
Files.readAllBytes(
Path.of("src/main/resources/cat.jpg")),
"image/jpeg")),
config);
for (Part part : response.parts()) {
if (part.text().isPresent()) {
System.out.println(part.text().get());
} else if (part.inlineData().isPresent()) {
var blob = part.inlineData().get();
if (blob.data().isPresent()) {
Files.write(Paths.get("gemini_generated_image.png"), blob.data().get());
}
}
}
}
}
}
REST
IMG_PATH=/path/to/cat_image.jpeg
if [[ "$(base64 --version 2>&1)" = *"FreeBSD"* ]]; then
B64FLAGS="--input"
else
B64FLAGS="-w0"
fi
IMG_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH" 2>&1)
curl -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-d "{
\"contents\": [{
\"parts\":[
{\"text\": \"'Create a picture of my cat eating a nano-banana in a fancy restaurant under the Gemini constellation\"},
{
\"inline_data\": {
\"mime_type\":\"image/jpeg\",
\"data\": \"$IMG_BASE64\"
}
}
]
}]
}" \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > gemini-edited-image.png
Wieloetapowa edycja obrazów
Kontynuuj generowanie i edytowanie obrazów w formie rozmowy. Czat lub rozmowa wieloetapowa to zalecany sposób na iteracyjne tworzenie obrazów. Poniższy przykład pokazuje prompta do wygenerowania infografiki na temat fotosyntezy.
Python
from google import genai
from google.genai import types
client = genai.Client()
chat = client.chats.create(
model="gemini-3-pro-image-preview",
config=types.GenerateContentConfig(
response_modalities=['TEXT', 'IMAGE'],
tools=[{"google_search": {}}]
)
)
message = "Create a vibrant infographic that explains photosynthesis as if it were a recipe for a plant's favorite food. Show the \"ingredients\" (sunlight, water, CO2) and the \"finished dish\" (sugar/energy). The style should be like a page from a colorful kids' cookbook, suitable for a 4th grader."
response = chat.send_message(message)
for part in response.parts:
if part.text is not None:
print(part.text)
elif image:= part.as_image():
image.save("photosynthesis.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
const ai = new GoogleGenAI({});
async function main() {
const chat = ai.chats.create({
model: "gemini-3-pro-image-preview",
config: {
responseModalities: ['TEXT', 'IMAGE'],
tools: [{googleSearch: {}}],
},
});
await main();
const message = "Create a vibrant infographic that explains photosynthesis as if it were a recipe for a plant's favorite food. Show the \"ingredients\" (sunlight, water, CO2) and the \"finished dish\" (sugar/energy). The style should be like a page from a colorful kids' cookbook, suitable for a 4th grader."
let response = await chat.sendMessage({message});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("photosynthesis.png", buffer);
console.log("Image saved as photosynthesis.png");
}
}
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
defer client.Close()
model := client.GenerativeModel("gemini-3-pro-image-preview")
model.GenerationConfig = &pb.GenerationConfig{
ResponseModalities: []pb.ResponseModality{genai.Text, genai.Image},
}
chat := model.StartChat()
message := "Create a vibrant infographic that explains photosynthesis as if it were a recipe for a plant's favorite food. Show the \"ingredients\" (sunlight, water, CO2) and the \"finished dish\" (sugar/energy). The style should be like a page from a colorful kids' cookbook, suitable for a 4th grader."
resp, err := chat.SendMessage(ctx, genai.Text(message))
if err != nil {
log.Fatal(err)
}
for _, part := range resp.Candidates[0].Content.Parts {
if txt, ok := part.(genai.Text); ok {
fmt.Printf("%s", string(txt))
} else if img, ok := part.(genai.ImageData); ok {
err := os.WriteFile("photosynthesis.png", img.Data, 0644)
if err != nil {
log.Fatal(err)
}
}
}
}
Java
import com.google.genai.Chat;
import com.google.genai.Client;
import com.google.genai.types.Content;
import com.google.genai.types.GenerateContentConfig;
import com.google.genai.types.GenerateContentResponse;
import com.google.genai.types.GoogleSearch;
import com.google.genai.types.ImageConfig;
import com.google.genai.types.Part;
import com.google.genai.types.RetrievalConfig;
import com.google.genai.types.Tool;
import com.google.genai.types.ToolConfig;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;
public class MultiturnImageEditing {
public static void main(String[] args) throws IOException {
try (Client client = new Client()) {
GenerateContentConfig config = GenerateContentConfig.builder()
.responseModalities("TEXT", "IMAGE")
.tools(Tool.builder()
.googleSearch(GoogleSearch.builder().build())
.build())
.build();
Chat chat = client.chats.create("gemini-3-pro-image-preview", config);
GenerateContentResponse response = chat.sendMessage("""
Create a vibrant infographic that explains photosynthesis
as if it were a recipe for a plant's favorite food.
Show the "ingredients" (sunlight, water, CO2)
and the "finished dish" (sugar/energy).
The style should be like a page from a colorful
kids' cookbook, suitable for a 4th grader.
""");
for (Part part : response.parts()) {
if (part.text().isPresent()) {
System.out.println(part.text().get());
} else if (part.inlineData().isPresent()) {
var blob = part.inlineData().get();
if (blob.data().isPresent()) {
Files.write(Paths.get("photosynthesis.png"), blob.data().get());
}
}
}
// ...
}
}
}
REST
curl -s -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-image-preview:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{
"role": "user",
"parts": [
{"text": "Create a vibrant infographic that explains photosynthesis as if it were a recipe for a plants favorite food. Show the \"ingredients\" (sunlight, water, CO2) and the \"finished dish\" (sugar/energy). The style should be like a page from a colorful kids cookbook, suitable for a 4th grader."}
]
}],
"generationConfig": {
"responseModalities": ["TEXT", "IMAGE"]
}
}' > turn1_response.json
cat turn1_response.json
# Requires jq to parse JSON response
jq -r '.candidates[0].content.parts[] | select(.inlineData) | .inlineData.data' turn1_response.json | head -1 | base64 --decode > photosynthesis.png
Następnie możesz użyć tego samego czatu, aby zmienić język na grafice na hiszpański.
Python
message = "Update this infographic to be in Spanish. Do not change any other elements of the image."
aspect_ratio = "16:9" # "1:1","2:3","3:2","3:4","4:3","4:5","5:4","9:16","16:9","21:9"
resolution = "2K" # "1K", "2K", "4K"
response = chat.send_message(message,
config=types.GenerateContentConfig(
image_config=types.ImageConfig(
aspect_ratio=aspect_ratio,
image_size=resolution
),
))
for part in response.parts:
if part.text is not None:
print(part.text)
elif image:= part.as_image():
image.save("photosynthesis_spanish.png")
JavaScript
const message = 'Update this infographic to be in Spanish. Do not change any other elements of the image.';
const aspectRatio = '16:9';
const resolution = '2K';
let response = await chat.sendMessage({
message,
config: {
responseModalities: ['TEXT', 'IMAGE'],
imageConfig: {
aspectRatio: aspectRatio,
imageSize: resolution,
},
tools: [{googleSearch: {}}],
},
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("photosynthesis2.png", buffer);
console.log("Image saved as photosynthesis2.png");
}
}
Przeczytaj
message = "Update this infographic to be in Spanish. Do not change any other elements of the image."
aspect_ratio = "16:9" // "1:1","2:3","3:2","3:4","4:3","4:5","5:4","9:16","16:9","21:9"
resolution = "2K" // "1K", "2K", "4K"
model.GenerationConfig.ImageConfig = &pb.ImageConfig{
AspectRatio: aspect_ratio,
ImageSize: resolution,
}
resp, err = chat.SendMessage(ctx, genai.Text(message))
if err != nil {
log.Fatal(err)
}
for _, part := range resp.Candidates[0].Content.Parts {
if txt, ok := part.(genai.Text); ok {
fmt.Printf("%s", string(txt))
} else if img, ok := part.(genai.ImageData); ok {
err := os.WriteFile("photosynthesis_spanish.png", img.Data, 0644)
if err != nil {
log.Fatal(err)
}
}
}
Java
String aspectRatio = "16:9"; // "1:1","2:3","3:2","3:4","4:3","4:5","5:4","9:16","16:9","21:9"
String resolution = "2K"; // "1K", "2K", "4K"
config = GenerateContentConfig.builder()
.responseModalities("TEXT", "IMAGE")
.imageConfig(ImageConfig.builder()
.aspectRatio(aspectRatio)
.imageSize(resolution)
.build())
.build();
response = chat.sendMessage(
"Update this infographic to be in Spanish. " +
"Do not change any other elements of the image.",
config);
for (Part part : response.parts()) {
if (part.text().isPresent()) {
System.out.println(part.text().get());
} else if (part.inlineData().isPresent()) {
var blob = part.inlineData().get();
if (blob.data().isPresent()) {
Files.write(Paths.get("photosynthesis_spanish.png"), blob.data().get());
}
}
}
REST
# Create request2.json by combining history and new prompt
# Read model's previous response content directly into jq
jq --argjson user1 '{"role": "user", "parts": [{"text": "Create a vibrant infographic that explains photosynthesis as if it were a recipe for a plant'\''s favorite food. Show the \"ingredients\" (sunlight, water, CO2) and the \"finished dish\" (sugar/energy). The style should be like a page from a colorful kids'\'' cookbook, suitable for a 4th grader."}]}' \
--argjson user2 '{"role": "user", "parts": [{"text": "Update this infographic to be in Spanish. Do not change any other elements of the image."}]}' \
-f /dev/stdin turn1_response.json > request2.json <<'EOF_JQ_FILTER'
.candidates[0].content | {
"contents": [$user1, ., $user2],
"tools": [{"google_search": {}}],
"generationConfig": {
"responseModalities": ["TEXT", "IMAGE"],
"imageConfig": {
"aspectRatio": "16:9",
"imageSize": "2K"
}
}
}
EOF_JQ_FILTER
curl -s -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-image-preview:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d @request2.json > turn2_response.json
jq -r '.candidates[0].content.parts[] | select(.inlineData) | .inlineData.data' turn2_response.json | head -1 | base64 --decode > photosynthesis_spanish.png
Nowość w Gemini 3 Pro Image
Gemini 3 Pro Image (gemini-3-pro-image-preview) to najnowocześniejszy model do generowania i edytowania obrazów zoptymalizowany pod kątem profesjonalnego tworzenia zasobów.
Został zaprojektowany do obsługi najbardziej wymagających procesów dzięki zaawansowanemu rozumowaniu. Doskonale radzi sobie ze złożonymi zadaniami tworzenia i modyfikowania, które wymagają wielu interakcji.
- Wyjście w wysokiej rozdzielczości: wbudowane funkcje generowania obrazów w rozdzielczości 1K, 2K i 4K.
- Zaawansowane renderowanie tekstu: umożliwia generowanie czytelnego, stylizowanego tekstu do infografik, menu, diagramów i materiałów marketingowych.
- Grounding z użyciem wyszukiwarki Google: model może używać wyszukiwarki Google jako narzędzia do weryfikowania faktów i generowania obrazów na podstawie danych w czasie rzeczywistym (np. aktualnych map pogody, wykresów akcji, ostatnich wydarzeń).
- Tryb myślenia: model wykorzystuje proces „myślenia”, aby analizować złożone prompty. Generuje tymczasowe „obrazy myśli” (widoczne w backendzie, ale nie są naliczane), aby dopracować kompozycję przed wygenerowaniem końcowego obrazu wysokiej jakości.
- Do 14 obrazów referencyjnych: możesz teraz łączyć maksymalnie 14 obrazów referencyjnych, aby uzyskać obraz końcowy.
Używanie maksymalnie 14 obrazów referencyjnych
Gemini 3 Pro (wersja testowa) umożliwia mieszanie maksymalnie 14 obrazów referencyjnych. Te 14 obrazów może obejmować:
- Do 6 obrazów obiektów o wysokiej wierności, które mają zostać uwzględnione na obrazie końcowym
Maksymalnie 5 zdjęć osób, aby zachować spójność postaci
Python
from google import genai
from google.genai import types
from PIL import Image
prompt = "An office group photo of these people, they are making funny faces."
aspect_ratio = "5:4" # "1:1","2:3","3:2","3:4","4:3","4:5","5:4","9:16","16:9","21:9"
resolution = "2K" # "1K", "2K", "4K"
client = genai.Client()
response = client.models.generate_content(
model="gemini-3-pro-image-preview",
contents=[
prompt,
Image.open('person1.png'),
Image.open('person2.png'),
Image.open('person3.png'),
Image.open('person4.png'),
Image.open('person5.png'),
],
config=types.GenerateContentConfig(
response_modalities=['TEXT', 'IMAGE'],
image_config=types.ImageConfig(
aspect_ratio=aspect_ratio,
image_size=resolution
),
)
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif image:= part.as_image():
image.save("office.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
'An office group photo of these people, they are making funny faces.';
const aspectRatio = '5:4';
const resolution = '2K';
const contents = [
{ text: prompt },
{
inlineData: {
mimeType: "image/jpeg",
data: base64ImageFile1,
},
},
{
inlineData: {
mimeType: "image/jpeg",
data: base64ImageFile2,
},
},
{
inlineData: {
mimeType: "image/jpeg",
data: base64ImageFile3,
},
},
{
inlineData: {
mimeType: "image/jpeg",
data: base64ImageFile4,
},
},
{
inlineData: {
mimeType: "image/jpeg",
data: base64ImageFile5,
},
}
];
const response = await ai.models.generateContent({
model: 'gemini-3-pro-image-preview',
contents: contents,
config: {
responseModalities: ['TEXT', 'IMAGE'],
imageConfig: {
aspectRatio: aspectRatio,
imageSize: resolution,
},
},
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("image.png", buffer);
console.log("Image saved as image.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
defer client.Close()
model := client.GenerativeModel("gemini-3-pro-image-preview")
model.GenerationConfig = &pb.GenerationConfig{
ResponseModalities: []pb.ResponseModality{genai.Text, genai.Image},
ImageConfig: &pb.ImageConfig{
AspectRatio: "5:4",
ImageSize: "2K",
},
}
img1, err := os.ReadFile("person1.png")
if err != nil { log.Fatal(err) }
img2, err := os.ReadFile("person2.png")
if err != nil { log.Fatal(err) }
img3, err := os.ReadFile("person3.png")
if err != nil { log.Fatal(err) }
img4, err := os.ReadFile("person4.png")
if err != nil { log.Fatal(err) }
img5, err := os.ReadFile("person5.png")
if err != nil { log.Fatal(err) }
parts := []genai.Part{
genai.Text("An office group photo of these people, they are making funny faces."),
genai.ImageData{MIMEType: "image/png", Data: img1},
genai.ImageData{MIMEType: "image/png", Data: img2},
genai.ImageData{MIMEType: "image/png", Data: img3},
genai.ImageData{MIMEType: "image/png", Data: img4},
genai.ImageData{MIMEType: "image/png", Data: img5},
}
resp, err := model.GenerateContent(ctx, parts...)
if err != nil {
log.Fatal(err)
}
for _, part := range resp.Candidates[0].Content.Parts {
if txt, ok := part.(genai.Text); ok {
fmt.Printf("%s", string(txt))
} else if img, ok := part.(genai.ImageData); ok {
err := os.WriteFile("office.png", img.Data, 0644)
if err != nil {
log.Fatal(err)
}
}
}
}
Java
import com.google.genai.Client;
import com.google.genai.types.Content;
import com.google.genai.types.GenerateContentConfig;
import com.google.genai.types.GenerateContentResponse;
import com.google.genai.types.ImageConfig;
import com.google.genai.types.Part;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;
public class GroupPhoto {
public static void main(String[] args) throws IOException {
try (Client client = new Client()) {
GenerateContentConfig config = GenerateContentConfig.builder()
.responseModalities("TEXT", "IMAGE")
.imageConfig(ImageConfig.builder()
.aspectRatio("5:4")
.imageSize("2K")
.build())
.build();
GenerateContentResponse response = client.models.generateContent(
"gemini-3-pro-image-preview",
Content.fromParts(
Part.fromText("An office group photo of these people, they are making funny faces."),
Part.fromBytes(Files.readAllBytes(Path.of("person1.png")), "image/png"),
Part.fromBytes(Files.readAllBytes(Path.of("person2.png")), "image/png"),
Part.fromBytes(Files.readAllBytes(Path.of("person3.png")), "image/png"),
Part.fromBytes(Files.readAllBytes(Path.of("person4.png")), "image/png"),
Part.fromBytes(Files.readAllBytes(Path.of("person5.png")), "image/png")
), config);
for (Part part : response.parts()) {
if (part.text().isPresent()) {
System.out.println(part.text().get());
} else if (part.inlineData().isPresent()) {
var blob = part.inlineData().get();
if (blob.data().isPresent()) {
Files.write(Paths.get("office.png"), blob.data().get());
}
}
}
}
}
}
REST
IMG_PATH1=person1.png
IMG_PATH2=person2.png
IMG_PATH3=person3.png
IMG_PATH4=person4.png
IMG_PATH5=person5.png
if [[ "$(base64 --version 2>&1)" = *"FreeBSD"* ]]; then
B64FLAGS="--input"
else
B64FLAGS="-w0"
fi
IMG1_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH1" 2>&1)
IMG2_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH2" 2>&1)
IMG3_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH3" 2>&1)
IMG4_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH4" 2>&1)
IMG5_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH5" 2>&1)
curl -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-image-preview:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-d "{
\"contents\": [{
\"parts\":[
{\"text\": \"An office group photo of these people, they are making funny faces.\"},
{\"inline_data\": {\"mime_type\":\"image/png\", \"data\": \"$IMG1_BASE64\"}},
{\"inline_data\": {\"mime_type\":\"image/png\", \"data\": \"$IMG2_BASE64\"}},
{\"inline_data\": {\"mime_type\":\"image/png\", \"data\": \"$IMG3_BASE64\"}},
{\"inline_data\": {\"mime_type\":\"image/png\", \"data\": \"$IMG4_BASE64\"}},
{\"inline_data\": {\"mime_type\":\"image/png\", \"data\": \"$IMG5_BASE64\"}}
]
}],
\"generationConfig\": {
\"responseModalities\": [\"TEXT\", \"IMAGE\"],
\"imageConfig\": {
\"aspectRatio\": \"5:4\",
\"imageSize\": \"2K\"
}
}
}" | jq -r '.candidates[0].content.parts[] | select(.inlineData) | .inlineData.data' | head -1 | base64 --decode > office.png
Powiązanie ze źródłem informacji przy użyciu wyszukiwarki Google
Użyj narzędzia wyszukiwarki Google, aby generować obrazy na podstawie informacji w czasie rzeczywistym, takich jak prognozy pogody, wykresy akcji czy ostatnie wydarzenia.
Uwagi, o których należy pamiętać podczas korzystania z powiązania ze źródłem informacji przy użyciu wyszukiwarki Google z generowaniem obrazów:
- Wyniki wyszukiwania oparte na obrazach nie są przekazywane do modelu generowania i są wykluczane z odpowiedzi.
Tryb tylko obraz (
responseModalities = ["IMAGE"]) nie zwraca obrazu, gdy jest używany z groundingiem przy użyciu wyszukiwarki Google.
Python
from google import genai
prompt = "Visualize the current weather forecast for the next 5 days in San Francisco as a clean, modern weather chart. Add a visual on what I should wear each day"
aspect_ratio = "16:9" # "1:1","2:3","3:2","3:4","4:3","4:5","5:4","9:16","16:9","21:9"
client = genai.Client()
response = client.models.generate_content(
model="gemini-3-pro-image-preview",
contents=prompt,
config=types.GenerateContentConfig(
response_modalities=['Text', 'Image'],
image_config=types.ImageConfig(
aspect_ratio=aspect_ratio,
),
tools=[{"google_search": {}}]
)
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif image:= part.as_image():
image.save("weather.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt = 'Visualize the current weather forecast for the next 5 days in San Francisco as a clean, modern weather chart. Add a visual on what I should wear each day';
const aspectRatio = '16:9';
const resolution = '2K';
const response = await ai.models.generateContent({
model: 'gemini-3-pro-image-preview',
contents: prompt,
config: {
responseModalities: ['TEXT', 'IMAGE'],
imageConfig: {
aspectRatio: aspectRatio,
imageSize: resolution,
},
},
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("image.png", buffer);
console.log("Image saved as image.png");
}
}
}
main();
Java
import com.google.genai.Client;
import com.google.genai.types.GenerateContentConfig;
import com.google.genai.types.GenerateContentResponse;
import com.google.genai.types.GoogleSearch;
import com.google.genai.types.ImageConfig;
import com.google.genai.types.Part;
import com.google.genai.types.Tool;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Paths;
public class SearchGrounding {
public static void main(String[] args) throws IOException {
try (Client client = new Client()) {
GenerateContentConfig config = GenerateContentConfig.builder()
.responseModalities("TEXT", "IMAGE")
.imageConfig(ImageConfig.builder()
.aspectRatio("16:9")
.build())
.tools(Tool.builder()
.googleSearch(GoogleSearch.builder().build())
.build())
.build();
GenerateContentResponse response = client.models.generateContent(
"gemini-3-pro-image-preview", """
Visualize the current weather forecast for the next 5 days
in San Francisco as a clean, modern weather chart.
Add a visual on what I should wear each day
""",
config);
for (Part part : response.parts()) {
if (part.text().isPresent()) {
System.out.println(part.text().get());
} else if (part.inlineData().isPresent()) {
var blob = part.inlineData().get();
if (blob.data().isPresent()) {
Files.write(Paths.get("weather.png"), blob.data().get());
}
}
}
}
}
}
REST
curl -s -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-image-preview:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{"parts": [{"text": "Visualize the current weather forecast for the next 5 days in San Francisco as a clean, modern weather chart. Add a visual on what I should wear each day"}]}],
"tools": [{"google_search": {}}],
"generationConfig": {
"responseModalities": ["TEXT", "IMAGE"],
"imageConfig": {"aspectRatio": "16:9"}
}
}' | jq -r '.candidates[0].content.parts[] | select(.inlineData) | .inlineData.data' | head -1 | base64 --decode > weather.png
Odpowiedź zawiera groundingMetadata, które zawiera te wymagane pola:
searchEntryPoint: zawiera kod HTML i CSS do renderowania wymaganych sugestii wyszukiwania.groundingChunks: zwraca 3 najpopularniejsze źródła internetowe użyte do utworzenia obrazu.
Generowanie obrazów w rozdzielczości do 4K
Model Gemini 3 Pro Image domyślnie generuje obrazy w rozdzielczości 1K, ale może też tworzyć obrazy w rozdzielczości 2K i 4K. Aby wygenerować zasoby w wyższej rozdzielczości, w parametrze generation_config podaj wartość image_size.
Musisz użyć wielkiej litery „K” (np. 1K, 2K, 4K). Parametry pisane małą literą (np. 1k) zostanie odrzucona.
Python
from google import genai
from google.genai import types
prompt = "Da Vinci style anatomical sketch of a dissected Monarch butterfly. Detailed drawings of the head, wings, and legs on textured parchment with notes in English."
aspect_ratio = "1:1" # "1:1","2:3","3:2","3:4","4:3","4:5","5:4","9:16","16:9","21:9"
resolution = "1K" # "1K", "2K", "4K"
client = genai.Client()
response = client.models.generate_content(
model="gemini-3-pro-image-preview",
contents=prompt,
config=types.GenerateContentConfig(
response_modalities=['TEXT', 'IMAGE'],
image_config=types.ImageConfig(
aspect_ratio=aspect_ratio,
image_size=resolution
),
)
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif image:= part.as_image():
image.save("butterfly.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
'Da Vinci style anatomical sketch of a dissected Monarch butterfly. Detailed drawings of the head, wings, and legs on textured parchment with notes in English.';
const aspectRatio = '1:1';
const resolution = '1K';
const response = await ai.models.generateContent({
model: 'gemini-3-pro-image-preview',
contents: prompt,
config: {
responseModalities: ['TEXT', 'IMAGE'],
imageConfig: {
aspectRatio: aspectRatio,
imageSize: resolution,
},
},
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("image.png", buffer);
console.log("Image saved as image.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
defer client.Close()
model := client.GenerativeModel("gemini-3-pro-image-preview")
model.GenerationConfig = &pb.GenerationConfig{
ResponseModalities: []pb.ResponseModality{genai.Text, genai.Image},
ImageConfig: &pb.ImageConfig{
AspectRatio: "1:1",
ImageSize: "1K",
},
}
prompt := "Da Vinci style anatomical sketch of a dissected Monarch butterfly. Detailed drawings of the head, wings, and legs on textured parchment with notes in English."
resp, err := model.GenerateContent(ctx, genai.Text(prompt))
if err != nil {
log.Fatal(err)
}
for _, part := range resp.Candidates[0].Content.Parts {
if txt, ok := part.(genai.Text); ok {
fmt.Printf("%s", string(txt))
} else if img, ok := part.(genai.ImageData); ok {
err := os.WriteFile("butterfly.png", img.Data, 0644)
if err != nil {
log.Fatal(err)
}
}
}
}
Java
import com.google.genai.Client;
import com.google.genai.types.GenerateContentConfig;
import com.google.genai.types.GenerateContentResponse;
import com.google.genai.types.GoogleSearch;
import com.google.genai.types.ImageConfig;
import com.google.genai.types.Part;
import com.google.genai.types.Tool;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Paths;
public class HiRes {
public static void main(String[] args) throws IOException {
try (Client client = new Client()) {
GenerateContentConfig config = GenerateContentConfig.builder()
.responseModalities("TEXT", "IMAGE")
.imageConfig(ImageConfig.builder()
.aspectRatio("16:9")
.imageSize("4K")
.build())
.build();
GenerateContentResponse response = client.models.generateContent(
"gemini-3-pro-image-preview", """
Da Vinci style anatomical sketch of a dissected Monarch butterfly.
Detailed drawings of the head, wings, and legs on textured
parchment with notes in English.
""",
config);
for (Part part : response.parts()) {
if (part.text().isPresent()) {
System.out.println(part.text().get());
} else if (part.inlineData().isPresent()) {
var blob = part.inlineData().get();
if (blob.data().isPresent()) {
Files.write(Paths.get("butterfly.png"), blob.data().get());
}
}
}
}
}
}
REST
curl -s -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-image-preview:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{"parts": [{"text": "Da Vinci style anatomical sketch of a dissected Monarch butterfly. Detailed drawings of the head, wings, and legs on textured parchment with notes in English."}]}],
"tools": [{"google_search": {}}],
"generationConfig": {
"responseModalities": ["TEXT", "IMAGE"],
"imageConfig": {"aspectRatio": "1:1", "imageSize": "1K"}
}
}' | jq -r '.candidates[0].content.parts[] | select(.inlineData) | .inlineData.data' | head -1 | base64 --decode > butterfly.png
Oto przykładowy obraz wygenerowany na podstawie tego prompta:
Proces myślowy
Model Gemini 3 Pro Image Preview to model myślący, który w przypadku złożonych promptów wykorzystuje proces rozumowania („Myślenie”). Ta funkcja jest domyślnie włączona i nie można jej wyłączyć w interfejsie API. Więcej informacji o procesie myślowym znajdziesz w przewodniku Myślenie Gemini.
Model generuje maksymalnie 2 obrazy tymczasowe, aby przetestować kompozycję i logikę. Ostatni obraz w sekcji „Myślenie” jest też ostatecznym wyrenderowanym obrazem.
Możesz sprawdzić, jakie myśli doprowadziły do wygenerowania ostatecznego obrazu.
Python
for part in response.parts:
if part.thought:
if part.text:
print(part.text)
elif image:= part.as_image():
image.show()
JavaScript
for (const part of response.candidates[0].content.parts) {
if (part.thought) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, 'base64');
fs.writeFileSync('image.png', buffer);
console.log('Image saved as image.png');
}
}
}
Podpisy myśli
Podpisy myśli to zaszyfrowane reprezentacje wewnętrznego procesu myślowego modelu. Służą one do zachowania kontekstu rozumowania w interakcjach wieloetapowych. Wszystkie odpowiedzi zawierają pole thought_signature. Ogólnie rzecz biorąc, jeśli w odpowiedzi modelu otrzymasz podpis myśli, w kolejnej turze należy go przekazać w niezmienionej formie wraz z historią rozmowy. Brak rozpowszechniania sygnatur myśli
może spowodować niepowodzenie odpowiedzi. Więcej informacji o sygnaturach znajdziesz w dokumentacji sygnatury myśli.
Oto jak działają sygnatury myśli:
- Wszystkie części
inline_dataz obrazemmimetype, które są częścią odpowiedzi, powinny mieć podpis. - Jeśli na początku (przed jakimkolwiek obrazem) zaraz po przemyśleniach znajdują się fragmenty tekstu, pierwszy z nich również powinien zawierać podpis.
- Myśli nie mają podpisów. Jeśli
inline_dataczęści z obrazemmimetypesą częścią myśli, nie będą miały podpisów.
Poniższy kod pokazuje przykład miejsca, w którym znajdują się sygnatury myśli:
[
{
"inline_data": {
"data": "<base64_image_data_0>",
"mime_type": "image/png"
},
"thought": true // Thoughts don't have signatures
},
{
"inline_data": {
"data": "<base64_image_data_1>",
"mime_type": "image/png"
},
"thought": true // Thoughts don't have signatures
},
{
"inline_data": {
"data": "<base64_image_data_2>",
"mime_type": "image/png"
},
"thought": true // Thoughts don't have signatures
},
{
"text": "Here is a step-by-step guide to baking macarons, presented in three separate images.\n\n### Step 1: Piping the Batter\n\nThe first step after making your macaron batter is to pipe it onto a baking sheet. This requires a steady hand to create uniform circles.\n\n",
"thought_signature": "<Signature_A>" // The first non-thought part always has a signature
},
{
"inline_data": {
"data": "<base64_image_data_3>",
"mime_type": "image/png"
},
"thought_signature": "<Signature_B>" // All image parts have a signatures
},
{
"text": "\n\n### Step 2: Baking and Developing Feet\n\nOnce piped, the macarons are baked in the oven. A key sign of a successful bake is the development of \"feet\"—the ruffled edge at the base of each macaron shell.\n\n"
// Follow-up text parts don't have signatures
},
{
"inline_data": {
"data": "<base64_image_data_4>",
"mime_type": "image/png"
},
"thought_signature": "<Signature_C>" // All image parts have a signatures
},
{
"text": "\n\n### Step 3: Assembling the Macaron\n\nThe final step is to pair the cooled macaron shells by size and sandwich them together with your desired filling, creating the classic macaron dessert.\n\n"
},
{
"inline_data": {
"data": "<base64_image_data_5>",
"mime_type": "image/png"
},
"thought_signature": "<Signature_D>" // All image parts have a signatures
}
]
Inne tryby generowania obrazów
Gemini obsługuje inne tryby interakcji z obrazami w zależności od struktury promptu i kontekstu, w tym:
- Tekst na obraz(y) i tekst (przeplatany): generuje obrazy z powiązanym tekstem.
- Przykładowy prompt: „Wygeneruj ilustrowany przepis na paellę”.
- Obrazy i tekst na obrazy i tekst (przeplatane): wykorzystuje obrazy i tekst wejściowy do tworzenia nowych powiązanych obrazów i tekstu.
- Przykładowy prompt: (z obrazem umeblowanego pokoju) „Jakie inne kolory sof pasowałyby do tego pomieszczenia? Czy możesz zaktualizować obraz?”.
Przewodnik po promptach i strategiach
Opanowanie generowania obrazów zaczyna się od jednej podstawowej zasady:
Opisz scenę, nie podawaj tylko słów kluczowych. Główną zaletą modelu jest dogłębne rozumienie języka. Opis w formie narracji, czyli opisowy akapit, prawie zawsze da lepszy i bardziej spójny obraz niż lista niepowiązanych ze sobą słów.
Prompty do generowania obrazów
Poniższe strategie pomogą Ci tworzyć skuteczne prompty, aby generować dokładnie takie obrazy, jakich szukasz.
1. Fotorealistyczne sceny
W przypadku realistycznych obrazów używaj terminów fotograficznych. Wspomnij o kątach kamery, typach obiektywów, oświetleniu i drobnych szczegółach, aby skierować model w stronę fotorealistycznego wyniku.
Szablon
A photorealistic [shot type] of [subject], [action or expression], set in
[environment]. The scene is illuminated by [lighting description], creating
a [mood] atmosphere. Captured with a [camera/lens details], emphasizing
[key textures and details]. The image should be in a [aspect ratio] format.
Prompt
A photorealistic close-up portrait of an elderly Japanese ceramicist with
deep, sun-etched wrinkles and a warm, knowing smile. He is carefully
inspecting a freshly glazed tea bowl. The setting is his rustic,
sun-drenched workshop. The scene is illuminated by soft, golden hour light
streaming through a window, highlighting the fine texture of the clay.
Captured with an 85mm portrait lens, resulting in a soft, blurred background
(bokeh). The overall mood is serene and masterful. Vertical portrait
orientation.
Python
from google import genai
from google.genai import types
client = genai.Client()
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents="A photorealistic close-up portrait of an elderly Japanese ceramicist with deep, sun-etched wrinkles and a warm, knowing smile. He is carefully inspecting a freshly glazed tea bowl. The setting is his rustic, sun-drenched workshop with pottery wheels and shelves of clay pots in the background. The scene is illuminated by soft, golden hour light streaming through a window, highlighting the fine texture of the clay and the fabric of his apron. Captured with an 85mm portrait lens, resulting in a soft, blurred background (bokeh). The overall mood is serene and masterful.",
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("photorealistic_example.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
"A photorealistic close-up portrait of an elderly Japanese ceramicist with deep, sun-etched wrinkles and a warm, knowing smile. He is carefully inspecting a freshly glazed tea bowl. The setting is his rustic, sun-drenched workshop with pottery wheels and shelves of clay pots in the background. The scene is illuminated by soft, golden hour light streaming through a window, highlighting the fine texture of the clay and the fabric of his apron. Captured with an 85mm portrait lens, resulting in a soft, blurred background (bokeh). The overall mood is serene and masterful.";
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("photorealistic_example.png", buffer);
console.log("Image saved as photorealistic_example.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("A photorealistic close-up portrait of an elderly Japanese ceramicist with deep, sun-etched wrinkles and a warm, knowing smile. He is carefully inspecting a freshly glazed tea bowl. The setting is his rustic, sun-drenched workshop with pottery wheels and shelves of clay pots in the background. The scene is illuminated by soft, golden hour light streaming through a window, highlighting the fine texture of the clay and the fabric of his apron. Captured with an 85mm portrait lens, resulting in a soft, blurred background (bokeh). The overall mood is serene and masterful."),
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "photorealistic_example.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
curl -s -X POST
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{
"parts": [
{"text": "A photorealistic close-up portrait of an elderly Japanese ceramicist with deep, sun-etched wrinkles and a warm, knowing smile. He is carefully inspecting a freshly glazed tea bowl. The setting is his rustic, sun-drenched workshop with pottery wheels and shelves of clay pots in the background. The scene is illuminated by soft, golden hour light streaming through a window, highlighting the fine texture of the clay and the fabric of his apron. Captured with an 85mm portrait lens, resulting in a soft, blurred background (bokeh). The overall mood is serene and masterful."}
]
}]
}' \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > photorealistic_example.png
2. Stylizowane ilustracje i naklejki
Aby utworzyć naklejki, ikony lub komponenty, podaj konkretny styl i poproś o przezroczyste tło.
Szablon
A [style] sticker of a [subject], featuring [key characteristics] and a
[color palette]. The design should have [line style] and [shading style].
The background must be transparent.
Prompt
A kawaii-style sticker of a happy red panda wearing a tiny bamboo hat. It's
munching on a green bamboo leaf. The design features bold, clean outlines,
simple cel-shading, and a vibrant color palette. The background must be white.
Python
from google import genai
from google.genai import types
client = genai.Client()
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents="A kawaii-style sticker of a happy red panda wearing a tiny bamboo hat. It's munching on a green bamboo leaf. The design features bold, clean outlines, simple cel-shading, and a vibrant color palette. The background must be white.",
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("red_panda_sticker.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
"A kawaii-style sticker of a happy red panda wearing a tiny bamboo hat. It's munching on a green bamboo leaf. The design features bold, clean outlines, simple cel-shading, and a vibrant color palette. The background must be white.";
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("red_panda_sticker.png", buffer);
console.log("Image saved as red_panda_sticker.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("A kawaii-style sticker of a happy red panda wearing a tiny bamboo hat. It's munching on a green bamboo leaf. The design features bold, clean outlines, simple cel-shading, and a vibrant color palette. The background must be white."),
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "red_panda_sticker.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
curl -s -X POST
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{
"parts": [
{"text": "A kawaii-style sticker of a happy red panda wearing a tiny bamboo hat. It'"'"'s munching on a green bamboo leaf. The design features bold, clean outlines, simple cel-shading, and a vibrant color palette. The background must be white."}
]
}]
}' \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > red_panda_sticker.png
3. Dokładny tekst na obrazach
Gemini świetnie radzi sobie z renderowaniem tekstu. Opisz dokładnie tekst, styl czcionki i ogólny projekt.
Szablon
Create a [image type] for [brand/concept] with the text "[text to render]"
in a [font style]. The design should be [style description], with a
[color scheme].
Prompt
Create a modern, minimalist logo for a coffee shop called 'The Daily Grind'.
The text should be in a clean, bold, sans-serif font. The design should
feature a simple, stylized icon of a a coffee bean seamlessly integrated
with the text. The color scheme is black and white.
Python
from google import genai
from google.genai import types
client = genai.Client()
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents="Create a modern, minimalist logo for a coffee shop called 'The Daily Grind'. The text should be in a clean, bold, sans-serif font. The design should feature a simple, stylized icon of a a coffee bean seamlessly integrated with the text. The color scheme is black and white.",
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("logo_example.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
"Create a modern, minimalist logo for a coffee shop called 'The Daily Grind'. The text should be in a clean, bold, sans-serif font. The design should feature a simple, stylized icon of a a coffee bean seamlessly integrated with the text. The color scheme is black and white.";
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("logo_example.png", buffer);
console.log("Image saved as logo_example.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("Create a modern, minimalist logo for a coffee shop called 'The Daily Grind'. The text should be in a clean, bold, sans-serif font. The design should feature a simple, stylized icon of a a coffee bean seamlessly integrated with the text. The color scheme is black and white."),
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "logo_example.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
curl -s -X POST
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{
"parts": [
{"text": "Create a modern, minimalist logo for a coffee shop called '"'"'The Daily Grind'"'"'. The text should be in a clean, bold, sans-serif font. The design should feature a simple, stylized icon of a a coffee bean seamlessly integrated with the text. The color scheme is black and white."}
]
}]
}' \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > logo_example.png
4. Makiety produktów i fotografia komercyjna
Idealne do tworzenia czystych, profesjonalnych zdjęć produktów na potrzeby e-commerce, reklam lub brandingu.
Szablon
A high-resolution, studio-lit product photograph of a [product description]
on a [background surface/description]. The lighting is a [lighting setup,
e.g., three-point softbox setup] to [lighting purpose]. The camera angle is
a [angle type] to showcase [specific feature]. Ultra-realistic, with sharp
focus on [key detail]. [Aspect ratio].
Prompt
A high-resolution, studio-lit product photograph of a minimalist ceramic
coffee mug in matte black, presented on a polished concrete surface. The
lighting is a three-point softbox setup designed to create soft, diffused
highlights and eliminate harsh shadows. The camera angle is a slightly
elevated 45-degree shot to showcase its clean lines. Ultra-realistic, with
sharp focus on the steam rising from the coffee. Square image.
Python
from google import genai
from google.genai import types
client = genai.Client()
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents="A high-resolution, studio-lit product photograph of a minimalist ceramic coffee mug in matte black, presented on a polished concrete surface. The lighting is a three-point softbox setup designed to create soft, diffused highlights and eliminate harsh shadows. The camera angle is a slightly elevated 45-degree shot to showcase its clean lines. Ultra-realistic, with sharp focus on the steam rising from the coffee. Square image.",
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("product_mockup.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
"A high-resolution, studio-lit product photograph of a minimalist ceramic coffee mug in matte black, presented on a polished concrete surface. The lighting is a three-point softbox setup designed to create soft, diffused highlights and eliminate harsh shadows. The camera angle is a slightly elevated 45-degree shot to showcase its clean lines. Ultra-realistic, with sharp focus on the steam rising from the coffee. Square image.";
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("product_mockup.png", buffer);
console.log("Image saved as product_mockup.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("A high-resolution, studio-lit product photograph of a minimalist ceramic coffee mug in matte black, presented on a polished concrete surface. The lighting is a three-point softbox setup designed to create soft, diffused highlights and eliminate harsh shadows. The camera angle is a slightly elevated 45-degree shot to showcase its clean lines. Ultra-realistic, with sharp focus on the steam rising from the coffee. Square image."),
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "product_mockup.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
curl -s -X POST
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{
"parts": [
{"text": "A high-resolution, studio-lit product photograph of a minimalist ceramic coffee mug in matte black, presented on a polished concrete surface. The lighting is a three-point softbox setup designed to create soft, diffused highlights and eliminate harsh shadows. The camera angle is a slightly elevated 45-degree shot to showcase its clean lines. Ultra-realistic, with sharp focus on the steam rising from the coffee. Square image."}
]
}]
}' \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > product_mockup.png
5. Minimalistyczny projekt z negatywną przestrzenią
Doskonale nadaje się do tworzenia tła stron internetowych, prezentacji lub materiałów marketingowych, na których będzie wyświetlany tekst.
Szablon
A minimalist composition featuring a single [subject] positioned in the
[bottom-right/top-left/etc.] of the frame. The background is a vast, empty
[color] canvas, creating significant negative space. Soft, subtle lighting.
[Aspect ratio].
Prompt
A minimalist composition featuring a single, delicate red maple leaf
positioned in the bottom-right of the frame. The background is a vast, empty
off-white canvas, creating significant negative space for text. Soft,
diffused lighting from the top left. Square image.
Python
from google import genai
from google.genai import types
client = genai.Client()
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents="A minimalist composition featuring a single, delicate red maple leaf positioned in the bottom-right of the frame. The background is a vast, empty off-white canvas, creating significant negative space for text. Soft, diffused lighting from the top left. Square image.",
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("minimalist_design.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
"A minimalist composition featuring a single, delicate red maple leaf positioned in the bottom-right of the frame. The background is a vast, empty off-white canvas, creating significant negative space for text. Soft, diffused lighting from the top left. Square image.";
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("minimalist_design.png", buffer);
console.log("Image saved as minimalist_design.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("A minimalist composition featuring a single, delicate red maple leaf positioned in the bottom-right of the frame. The background is a vast, empty off-white canvas, creating significant negative space for text. Soft, diffused lighting from the top left. Square image."),
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "minimalist_design.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
curl -s -X POST
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{
"parts": [
{"text": "A minimalist composition featuring a single, delicate red maple leaf positioned in the bottom-right of the frame. The background is a vast, empty off-white canvas, creating significant negative space for text. Soft, diffused lighting from the top left. Square image."}
]
}]
}' \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > minimalist_design.png
6. Sztuka sekwencyjna (panel komiksu / scenorys)
Opiera się na spójności postaci i opisie sceny, aby tworzyć panele do wizualnego opowiadania historii.
Szablon
A single comic book panel in a [art style] style. In the foreground,
[character description and action]. In the background, [setting details].
The panel has a [dialogue/caption box] with the text "[Text]". The lighting
creates a [mood] mood. [Aspect ratio].
Prompt
A single comic book panel in a gritty, noir art style with high-contrast
black and white inks. In the foreground, a detective in a trench coat stands
under a flickering streetlamp, rain soaking his shoulders. In the
background, the neon sign of a desolate bar reflects in a puddle. A caption
box at the top reads "The city was a tough place to keep secrets." The
lighting is harsh, creating a dramatic, somber mood. Landscape.
Python
from google import genai
from google.genai import types
client = genai.Client()
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents="A single comic book panel in a gritty, noir art style with high-contrast black and white inks. In the foreground, a detective in a trench coat stands under a flickering streetlamp, rain soaking his shoulders. In the background, the neon sign of a desolate bar reflects in a puddle. A caption box at the top reads \"The city was a tough place to keep secrets.\" The lighting is harsh, creating a dramatic, somber mood. Landscape.",
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("comic_panel.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const prompt =
"A single comic book panel in a gritty, noir art style with high-contrast black and white inks. In the foreground, a detective in a trench coat stands under a flickering streetlamp, rain soaking his shoulders. In the background, the neon sign of a desolate bar reflects in a puddle. A caption box at the top reads \"The city was a tough place to keep secrets.\" The lighting is harsh, creating a dramatic, somber mood. Landscape.";
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("comic_panel.png", buffer);
console.log("Image saved as comic_panel.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("A single comic book panel in a gritty, noir art style with high-contrast black and white inks. In the foreground, a detective in a trench coat stands under a flickering streetlamp, rain soaking his shoulders. In the background, the neon sign of a desolate bar reflects in a puddle. A caption box at the top reads \"The city was a tough place to keep secrets.\" The lighting is harsh, creating a dramatic, somber mood. Landscape."),
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "comic_panel.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
curl -s -X POST
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"contents": [{
"parts": [
{"text": "A single comic book panel in a gritty, noir art style with high-contrast black and white inks. In the foreground, a detective in a trench coat stands under a flickering streetlamp, rain soaking his shoulders. In the background, the neon sign of a desolate bar reflects in a puddle. A caption box at the top reads \"The city was a tough place to keep secrets.\" The lighting is harsh, creating a dramatic, somber mood. Landscape."}
]
}]
}' \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > comic_panel.png
Prompty do edytowania obrazów
Te przykłady pokazują, jak przesyłać obrazy wraz z promptami tekstowymi w celu edycji, kompozycji i przenoszenia stylu.
1. Dodawanie i usuwanie elementów
Prześlij obraz i opisz zmianę. Model będzie pasować do stylu, oświetlenia i perspektywy oryginalnego obrazu.
Szablon
Using the provided image of [subject], please [add/remove/modify] [element]
to/from the scene. Ensure the change is [description of how the change should
integrate].
Prompt
"Using the provided image of my cat, please add a small, knitted wizard hat
on its head. Make it look like it's sitting comfortably and matches the soft
lighting of the photo."
Python
from google import genai
from google.genai import types
from PIL import Image
client = genai.Client()
# Base image prompt: "A photorealistic picture of a fluffy ginger cat sitting on a wooden floor, looking directly at the camera. Soft, natural light from a window."
image_input = Image.open('/path/to/your/cat_photo.png')
text_input = """Using the provided image of my cat, please add a small, knitted wizard hat on its head. Make it look like it's sitting comfortably and not falling off."""
# Generate an image from a text prompt
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[text_input, image_input],
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("cat_with_hat.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const imagePath = "/path/to/your/cat_photo.png";
const imageData = fs.readFileSync(imagePath);
const base64Image = imageData.toString("base64");
const prompt = [
{ text: "Using the provided image of my cat, please add a small, knitted wizard hat on its head. Make it look like it's sitting comfortably and not falling off." },
{
inlineData: {
mimeType: "image/png",
data: base64Image,
},
},
];
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("cat_with_hat.png", buffer);
console.log("Image saved as cat_with_hat.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
imagePath := "/path/to/your/cat_photo.png"
imgData, _ := os.ReadFile(imagePath)
parts := []*genai.Part{
genai.NewPartFromText("Using the provided image of my cat, please add a small, knitted wizard hat on its head. Make it look like it's sitting comfortably and not falling off."),
&genai.Part{
InlineData: &genai.Blob{
MIMEType: "image/png",
Data: imgData,
},
},
}
contents := []*genai.Content{
genai.NewContentFromParts(parts, genai.RoleUser),
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
contents,
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "cat_with_hat.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
IMG_PATH=/path/to/your/cat_photo.png
if [[ "$(base64 --version 2>&1)" = *"FreeBSD"* ]]; then
B64FLAGS="--input"
else
B64FLAGS="-w0"
fi
IMG_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH" 2>&1)
curl -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-d "{
\"contents\": [{
\"parts\":[
{\"text\": \"Using the provided image of my cat, please add a small, knitted wizard hat on its head. Make it look like it's sitting comfortably and not falling off.\"},
{
\"inline_data\": {
\"mime_type\":\"image/png\",
\"data\": \"$IMG_BASE64\"
}
}
]
}]
}" \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > cat_with_hat.png
Dane wejściowe |
Wyniki |
|
|
2. Retusz (maskowanie semantyczne)
Określ „maskę” w rozmowie, aby edytować konkretną część obrazu, nie zmieniając reszty.
Szablon
Using the provided image, change only the [specific element] to [new
element/description]. Keep everything else in the image exactly the same,
preserving the original style, lighting, and composition.
Prompt
"Using the provided image of a living room, change only the blue sofa to be
a vintage, brown leather chesterfield sofa. Keep the rest of the room,
including the pillows on the sofa and the lighting, unchanged."
Python
from google import genai
from google.genai import types
from PIL import Image
client = genai.Client()
# Base image prompt: "A wide shot of a modern, well-lit living room with a prominent blue sofa in the center. A coffee table is in front of it and a large window is in the background."
living_room_image = Image.open('/path/to/your/living_room.png')
text_input = """Using the provided image of a living room, change only the blue sofa to be a vintage, brown leather chesterfield sofa. Keep the rest of the room, including the pillows on the sofa and the lighting, unchanged."""
# Generate an image from a text prompt
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[living_room_image, text_input],
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("living_room_edited.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const imagePath = "/path/to/your/living_room.png";
const imageData = fs.readFileSync(imagePath);
const base64Image = imageData.toString("base64");
const prompt = [
{
inlineData: {
mimeType: "image/png",
data: base64Image,
},
},
{ text: "Using the provided image of a living room, change only the blue sofa to be a vintage, brown leather chesterfield sofa. Keep the rest of the room, including the pillows on the sofa and the lighting, unchanged." },
];
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("living_room_edited.png", buffer);
console.log("Image saved as living_room_edited.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
imagePath := "/path/to/your/living_room.png"
imgData, _ := os.ReadFile(imagePath)
parts := []*genai.Part{
&genai.Part{
InlineData: &genai.Blob{
MIMEType: "image/png",
Data: imgData,
},
},
genai.NewPartFromText("Using the provided image of a living room, change only the blue sofa to be a vintage, brown leather chesterfield sofa. Keep the rest of the room, including the pillows on the sofa and the lighting, unchanged."),
}
contents := []*genai.Content{
genai.NewContentFromParts(parts, genai.RoleUser),
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
contents,
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "living_room_edited.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
IMG_PATH=/path/to/your/living_room.png
if [[ "$(base64 --version 2>&1)" = *"FreeBSD"* ]]; then
B64FLAGS="--input"
else
B64FLAGS="-w0"
fi
IMG_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH" 2>&1)
curl -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-d "{
\"contents\": [{
\"parts\":[
{
\"inline_data\": {
\"mime_type\":\"image/png\",
\"data\": \"$IMG_BASE64\"
}
},
{\"text\": \"Using the provided image of a living room, change only the blue sofa to be a vintage, brown leather chesterfield sofa. Keep the rest of the room, including the pillows on the sofa and the lighting, unchanged.\"}
]
}]
}" \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > living_room_edited.png
Dane wejściowe |
Wyniki |
|
|
3. Przenoszenie stylu
Prześlij obraz i poproś model o odtworzenie jego treści w innym stylu artystycznym.
Szablon
Transform the provided photograph of [subject] into the artistic style of [artist/art style]. Preserve the original composition but render it with [description of stylistic elements].
Prompt
"Transform the provided photograph of a modern city street at night into the artistic style of Vincent van Gogh's 'Starry Night'. Preserve the original composition of buildings and cars, but render all elements with swirling, impasto brushstrokes and a dramatic palette of deep blues and bright yellows."
Python
from google import genai
from google.genai import types
from PIL import Image
client = genai.Client()
# Base image prompt: "A photorealistic, high-resolution photograph of a busy city street in New York at night, with bright neon signs, yellow taxis, and tall skyscrapers."
city_image = Image.open('/path/to/your/city.png')
text_input = """Transform the provided photograph of a modern city street at night into the artistic style of Vincent van Gogh's 'Starry Night'. Preserve the original composition of buildings and cars, but render all elements with swirling, impasto brushstrokes and a dramatic palette of deep blues and bright yellows."""
# Generate an image from a text prompt
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[city_image, text_input],
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("city_style_transfer.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const imagePath = "/path/to/your/city.png";
const imageData = fs.readFileSync(imagePath);
const base64Image = imageData.toString("base64");
const prompt = [
{
inlineData: {
mimeType: "image/png",
data: base64Image,
},
},
{ text: "Transform the provided photograph of a modern city street at night into the artistic style of Vincent van Gogh's 'Starry Night'. Preserve the original composition of buildings and cars, but render all elements with swirling, impasto brushstrokes and a dramatic palette of deep blues and bright yellows." },
];
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("city_style_transfer.png", buffer);
console.log("Image saved as city_style_transfer.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
imagePath := "/path/to/your/city.png"
imgData, _ := os.ReadFile(imagePath)
parts := []*genai.Part{
&genai.Part{
InlineData: &genai.Blob{
MIMEType: "image/png",
Data: imgData,
},
},
genai.NewPartFromText("Transform the provided photograph of a modern city street at night into the artistic style of Vincent van Gogh's 'Starry Night'. Preserve the original composition of buildings and cars, but render all elements with swirling, impasto brushstrokes and a dramatic palette of deep blues and bright yellows."),
}
contents := []*genai.Content{
genai.NewContentFromParts(parts, genai.RoleUser),
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
contents,
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "city_style_transfer.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
IMG_PATH=/path/to/your/city.png
if [[ "$(base64 --version 2>&1)" = *"FreeBSD"* ]]; then
B64FLAGS="--input"
else
B64FLAGS="-w0"
fi
IMG_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH" 2>&1)
curl -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-d "{
\"contents\": [{
\"parts\":[
{
\"inline_data\": {
\"mime_type\":\"image/png\",
\"data\": \"$IMG_BASE64\"
}
},
{\"text\": \"Transform the provided photograph of a modern city street at night into the artistic style of Vincent van Gogh's 'Starry Night'. Preserve the original composition of buildings and cars, but render all elements with swirling, impasto brushstrokes and a dramatic palette of deep blues and bright yellows.\"}
]
}]
}" \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > city_style_transfer.png
Dane wejściowe |
Wyniki |
|
|
4. Zaawansowana kompozycja: łączenie wielu obrazów
Podaj kilka obrazów jako kontekst, aby utworzyć nową, złożoną scenę. To idealne rozwiązanie do tworzenia makiet produktów lub kreatywnych kolaży.
Szablon
Create a new image by combining the elements from the provided images. Take
the [element from image 1] and place it with/on the [element from image 2].
The final image should be a [description of the final scene].
Prompt
"Create a professional e-commerce fashion photo. Take the blue floral dress
from the first image and let the woman from the second image wear it.
Generate a realistic, full-body shot of the woman wearing the dress, with
the lighting and shadows adjusted to match the outdoor environment."
Python
from google import genai
from google.genai import types
from PIL import Image
client = genai.Client()
# Base image prompts:
# 1. Dress: "A professionally shot photo of a blue floral summer dress on a plain white background, ghost mannequin style."
# 2. Model: "Full-body shot of a woman with her hair in a bun, smiling, standing against a neutral grey studio background."
dress_image = Image.open('/path/to/your/dress.png')
model_image = Image.open('/path/to/your/model.png')
text_input = """Create a professional e-commerce fashion photo. Take the blue floral dress from the first image and let the woman from the second image wear it. Generate a realistic, full-body shot of the woman wearing the dress, with the lighting and shadows adjusted to match the outdoor environment."""
# Generate an image from a text prompt
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[dress_image, model_image, text_input],
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("fashion_ecommerce_shot.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const imagePath1 = "/path/to/your/dress.png";
const imageData1 = fs.readFileSync(imagePath1);
const base64Image1 = imageData1.toString("base64");
const imagePath2 = "/path/to/your/model.png";
const imageData2 = fs.readFileSync(imagePath2);
const base64Image2 = imageData2.toString("base64");
const prompt = [
{
inlineData: {
mimeType: "image/png",
data: base64Image1,
},
},
{
inlineData: {
mimeType: "image/png",
data: base64Image2,
},
},
{ text: "Create a professional e-commerce fashion photo. Take the blue floral dress from the first image and let the woman from the second image wear it. Generate a realistic, full-body shot of the woman wearing the dress, with the lighting and shadows adjusted to match the outdoor environment." },
];
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("fashion_ecommerce_shot.png", buffer);
console.log("Image saved as fashion_ecommerce_shot.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
imgData1, _ := os.ReadFile("/path/to/your/dress.png")
imgData2, _ := os.ReadFile("/path/to/your/model.png")
parts := []*genai.Part{
&genai.Part{
InlineData: &genai.Blob{
MIMEType: "image/png",
Data: imgData1,
},
},
&genai.Part{
InlineData: &genai.Blob{
MIMEType: "image/png",
Data: imgData2,
},
},
genai.NewPartFromText("Create a professional e-commerce fashion photo. Take the blue floral dress from the first image and let the woman from the second image wear it. Generate a realistic, full-body shot of the woman wearing the dress, with the lighting and shadows adjusted to match the outdoor environment."),
}
contents := []*genai.Content{
genai.NewContentFromParts(parts, genai.RoleUser),
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
contents,
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "fashion_ecommerce_shot.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
IMG_PATH1=/path/to/your/dress.png
IMG_PATH2=/path/to/your/model.png
if [[ "$(base64 --version 2>&1)" = *"FreeBSD"* ]]; then
B64FLAGS="--input"
else
B64FLAGS="-w0"
fi
IMG1_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH1" 2>&1)
IMG2_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH2" 2>&1)
curl -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-d "{
\"contents\": [{
\"parts\":[
{
\"inline_data\": {
\"mime_type\":\"image/png\",
\"data\": \"$IMG1_BASE64\"
}
},
{
\"inline_data\": {
\"mime_type\":\"image/png\",
\"data\": \"$IMG2_BASE64\"
}
},
{\"text\": \"Create a professional e-commerce fashion photo. Take the blue floral dress from the first image and let the woman from the second image wear it. Generate a realistic, full-body shot of the woman wearing the dress, with the lighting and shadows adjusted to match the outdoor environment.\"}
]
}]
}" \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > fashion_ecommerce_shot.png
Dane wejściowe 1 |
Dane wejściowe 2 |
Wyniki |
|
|
|
5. Zachowanie szczegółów w wysokiej jakości
Aby mieć pewność, że podczas edycji zostaną zachowane ważne szczegóły (np. twarz lub logo), dokładnie je opisz wraz z prośbą o edytowanie.
Szablon
Using the provided images, place [element from image 2] onto [element from
image 1]. Ensure that the features of [element from image 1] remain
completely unchanged. The added element should [description of how the
element should integrate].
Prompt
"Take the first image of the woman with brown hair, blue eyes, and a neutral
expression. Add the logo from the second image onto her black t-shirt.
Ensure the woman's face and features remain completely unchanged. The logo
should look like it's naturally printed on the fabric, following the folds
of the shirt."
Python
from google import genai
from google.genai import types
from PIL import Image
client = genai.Client()
# Base image prompts:
# 1. Woman: "A professional headshot of a woman with brown hair and blue eyes, wearing a plain black t-shirt, against a neutral studio background."
# 2. Logo: "A simple, modern logo with the letters 'G' and 'A' in a white circle."
woman_image = Image.open('/path/to/your/woman.png')
logo_image = Image.open('/path/to/your/logo.png')
text_input = """Take the first image of the woman with brown hair, blue eyes, and a neutral expression. Add the logo from the second image onto her black t-shirt. Ensure the woman's face and features remain completely unchanged. The logo should look like it's naturally printed on the fabric, following the folds of the shirt."""
# Generate an image from a text prompt
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[woman_image, logo_image, text_input],
)
for part in response.parts:
if part.text is not None:
print(part.text)
elif part.inline_data is not None:
image = part.as_image()
image.save("woman_with_logo.png")
JavaScript
import { GoogleGenAI } from "@google/genai";
import * as fs from "node:fs";
async function main() {
const ai = new GoogleGenAI({});
const imagePath1 = "/path/to/your/woman.png";
const imageData1 = fs.readFileSync(imagePath1);
const base64Image1 = imageData1.toString("base64");
const imagePath2 = "/path/to/your/logo.png";
const imageData2 = fs.readFileSync(imagePath2);
const base64Image2 = imageData2.toString("base64");
const prompt = [
{
inlineData: {
mimeType: "image/png",
data: base64Image1,
},
},
{
inlineData: {
mimeType: "image/png",
data: base64Image2,
},
},
{ text: "Take the first image of the woman with brown hair, blue eyes, and a neutral expression. Add the logo from the second image onto her black t-shirt. Ensure the woman's face and features remain completely unchanged. The logo should look like it's naturally printed on the fabric, following the folds of the shirt." },
];
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
});
for (const part of response.candidates[0].content.parts) {
if (part.text) {
console.log(part.text);
} else if (part.inlineData) {
const imageData = part.inlineData.data;
const buffer = Buffer.from(imageData, "base64");
fs.writeFileSync("woman_with_logo.png", buffer);
console.log("Image saved as woman_with_logo.png");
}
}
}
main();
Przeczytaj
package main
import (
"context"
"fmt"
"log"
"os"
"google.golang.org/genai"
)
func main() {
ctx := context.Background()
client, err := genai.NewClient(ctx, nil)
if err != nil {
log.Fatal(err)
}
imgData1, _ := os.ReadFile("/path/to/your/woman.png")
imgData2, _ := os.ReadFile("/path/to/your/logo.png")
parts := []*genai.Part{
&genai.Part{
InlineData: &genai.Blob{
MIMEType: "image/png",
Data: imgData1,
},
},
&genai.Part{
InlineData: &genai.Blob{
MIMEType: "image/png",
Data: imgData2,
},
},
genai.NewPartFromText("Take the first image of the woman with brown hair, blue eyes, and a neutral expression. Add the logo from the second image onto her black t-shirt. Ensure the woman's face and features remain completely unchanged. The logo should look like it's naturally printed on the fabric, following the folds of the shirt."),
}
contents := []*genai.Content{
genai.NewContentFromParts(parts, genai.RoleUser),
}
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
contents,
)
for _, part := range result.Candidates[0].Content.Parts {
if part.Text != "" {
fmt.Println(part.Text)
} else if part.InlineData != nil {
imageBytes := part.InlineData.Data
outputFilename := "woman_with_logo.png"
_ = os.WriteFile(outputFilename, imageBytes, 0644)
}
}
}
REST
IMG_PATH1=/path/to/your/woman.png
IMG_PATH2=/path/to/your/logo.png
if [[ "$(base64 --version 2>&1)" = *"FreeBSD"* ]]; then
B64FLAGS="--input"
else
B64FLAGS="-w0"
fi
IMG1_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH1" 2>&1)
IMG2_BASE64=$(base64 "$B64FLAGS" "$IMG_PATH2" 2>&1)
curl -X POST \
"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash-image:generateContent" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-d "{
\"contents\": [{
\"parts\":[
{
\"inline_data\": {
\"mime_type\":\"image/png\",
\"data\": \"$IMG1_BASE64\"
}
},
{
\"inline_data\": {
\"mime_type\":\"image/png\",
\"data\": \"$IMG2_BASE64\"
}
},
{\"text\": \"Take the first image of the woman with brown hair, blue eyes, and a neutral expression. Add the logo from the second image onto her black t-shirt. Ensure the woman's face and features remain completely unchanged. The logo should look like it's naturally printed on the fabric, following the folds of the shirt.\"}
]
}]
}" \
| grep -o '"data": "[^"]*"' \
| cut -d'"' -f4 \
| base64 --decode > woman_with_logo.png
Dane wejściowe 1 |
Dane wejściowe 2 |
Wyniki |
|
|
|
Sprawdzone metody
Aby poprawić wyniki, włącz do swojego przepływu pracy te profesjonalne strategie.
- Podawaj bardzo szczegółowe informacje: im więcej szczegółów podasz, tym większą będziesz mieć kontrolę. Zamiast „zbroja fantasy” opisz ją: „bogato zdobiona elfia zbroja płytowa, wyryte wzory z liści srebra, wysoki kołnierz i naramienniki w kształcie skrzydeł sokoła”.
- Podaj kontekst i intencję: wyjaśnij cel obrazu. Zrozumienie kontekstu przez model wpłynie na ostateczny wynik. Na przykład „Utwórz logo dla minimalistycznej marki kosmetyków do pielęgnacji skóry z wyższej półki” da lepsze wyniki niż samo „Utwórz logo”.
- Powtarzaj i dopracowuj: nie oczekuj idealnego obrazu za pierwszym razem. Wykorzystaj konwersacyjny charakter modelu, aby wprowadzać drobne zmiany. Następnie możesz użyć promptów takich jak „Świetnie, ale czy możesz trochę ocieplić oświetlenie?” lub „Pozostaw wszystko bez zmian, ale zmień wyraz twarzy postaci na bardziej poważny”.
- Używaj instrukcji krok po kroku: w przypadku złożonych scen z wieloma elementami podziel prompt na kroki. „Najpierw utwórz tło przedstawiające spokojny, zamglony las o świcie. Następnie na pierwszym planie dodaj pokryty mchem starożytny kamienny ołtarz. Na koniec połóż na ołtarzu jeden świecący miecz”.
- Używaj „semantycznych negatywnych promptów”: zamiast pisać „bez samochodów” opisz pożądaną scenę w sposób pozytywny: „pusta, opuszczona ulica bez śladów ruchu”.
- Sterowanie kamerą: używaj języka fotograficznego i filmowego, aby sterować kompozycją. Określenia takie jak
wide-angle shot,macro shot,low-angle perspective.
Ograniczenia
- Aby uzyskać najlepsze wyniki, używaj tych języków: EN, es-MX, ja-JP, zh-CN, hi-IN.
- Generowanie obrazów nie obsługuje danych wejściowych audio ani wideo.
- Model nie zawsze będzie generować dokładnie tyle obrazów, ile użytkownik wyraźnie poprosi.
- Model działa najlepiej, gdy jako dane wejściowe podasz maksymalnie 3 obrazy.
- Podczas generowania tekstu do obrazu Gemini działa najlepiej, jeśli najpierw wygenerujesz tekst, a potem poprosisz o obraz z tym tekstem.
- Wszystkie wygenerowane obrazy zawierają znak wodny SynthID.
Konfiguracje opcjonalne
Opcjonalnie możesz skonfigurować tryby odpowiedzi i format wyjściowy modelu w polu config wywołań generate_content.
Typy wyjściowe
Domyślnie model zwraca odpowiedzi tekstowe i graficzne (czyli response_modalities=['Text', 'Image']). Możesz skonfigurować odpowiedź tak, aby zwracała tylko obrazy bez tekstu, używając response_modalities=['Image'].
Python
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[prompt],
config=types.GenerateContentConfig(
response_modalities=['Image']
)
)
JavaScript
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
config: {
responseModalities: ['Image']
}
});
Przeczytaj
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("Create a picture of a nano banana dish in a " +
" fancy restaurant with a Gemini theme"),
&genai.GenerateContentConfig{
ResponseModalities: "Image",
},
)
REST
-d '{
"contents": [{
"parts": [
{"text": "Create a picture of a nano banana dish in a fancy restaurant with a Gemini theme"}
]
}],
"generationConfig": {
"responseModalities": ["Image"]
}
}' \
Formaty obrazu
Domyślnie model dopasowuje rozmiar obrazu wyjściowego do rozmiaru obrazu wejściowego lub generuje kwadraty o proporcjach 1:1.
Format obrazu wyjściowego możesz kontrolować za pomocą pola aspect_ratio w sekcji image_config w żądaniu odpowiedzi, jak pokazano poniżej:
Python
response = client.models.generate_content(
model="gemini-2.5-flash-image",
contents=[prompt],
config=types.GenerateContentConfig(
image_config=types.ImageConfig(
aspect_ratio="16:9",
)
)
)
JavaScript
const response = await ai.models.generateContent({
model: "gemini-2.5-flash-image",
contents: prompt,
config: {
imageConfig: {
aspectRatio: "16:9",
},
}
});
Przeczytaj
result, _ := client.Models.GenerateContent(
ctx,
"gemini-2.5-flash-image",
genai.Text("Create a picture of a nano banana dish in a " +
" fancy restaurant with a Gemini theme"),
&genai.GenerateContentConfig{
ImageConfig: &genai.ImageConfig{
AspectRatio: "16:9",
},
}
)
REST
-d '{
"contents": [{
"parts": [
{"text": "Create a picture of a nano banana dish in a fancy restaurant with a Gemini theme"}
]
}],
"generationConfig": {
"imageConfig": {
"aspectRatio": "16:9"
}
}
}' \
Dostępne proporcje i rozmiar wygenerowanego obrazu znajdziesz w tabelach poniżej:
Gemini 2.5 Flash Image
| Format obrazu | Rozdzielczość | Tokeny |
|---|---|---|
| 1:1 | 1024 × 1024 | 1290 |
| 2:3 | 832 x 1248 | 1290 |
| 3:2 | 1248 x 832 | 1290 |
| 3:4 | 864 x 1184 | 1290 |
| 4:3 | 1184 x 864 | 1290 |
| 4:5 | 896 x 1152 | 1290 |
| 5:4 | 1152 x 896 | 1290 |
| 9:16 | 768 x 1344 | 1290 |
| 16:9 | 1344 x 768 | 1290 |
| 21:9 | 1536 x 672 | 1290 |
Podgląd obrazu Gemini 3 Pro
| Format obrazu | Rozdzielczość 1K | 1 K tokenów | Rozdzielczość 2K | 2 tys. tokenów | Rozdzielczość 4K | 4K Tokens |
|---|---|---|---|---|---|---|
| 1:1 | 1024 × 1024 | 1210 | 2048 x 2048 | 1210 | 4096 x 4096 | 2000 |
| 2:3 | 848 x 1264 | 1210 | 1696 x 2528 | 1210 | 3392 x 5056 | 2000 |
| 3:2 | 1264 x 848 | 1210 | 2528 x 1696 | 1210 | 5056 x 3392 | 2000 |
| 3:4 | 896 x 1200 | 1210 | 1792 x 2400 | 1210 | 3584 x 4800 | 2000 |
| 4:3 | 1200 x 896 | 1210 | 2400 x 1792 | 1210 | 4800 x 3584 | 2000 |
| 4:5 | 928 x 1152 | 1210 | 1856 x 2304 | 1210 | 3712x4608 | 2000 |
| 5:4 | 1152 x 928 | 1210 | 2304 x 1856 | 1210 | 4608x3712 | 2000 |
| 9:16 | 768 x 1376 | 1210 | 1536 x 2752 | 1210 | 3072 x 5504 | 2000 |
| 16:9 | 1376 x 768 | 1210 | 2752 x 1536 | 1210 | 5504 x 3072 | 2000 |
| 21:9 | 1584 x 672 | 1210 | 3168 x 1344 | 1210 | 6336 x 2688 | 2000 |
Kiedy używać Imagen
Oprócz korzystania z wbudowanych funkcji generowania obrazów w Gemini możesz też uzyskać dostęp do Imagen, naszego specjalistycznego modelu generowania obrazów, za pomocą interfejsu Gemini API.
| Atrybut | Imagen | Obraz natywny Gemini |
|---|---|---|
| Mocne strony | Model specjalizuje się w generowaniu obrazów. | Domyślna rekomendacja. Niezrównana elastyczność, zrozumienie kontekstu oraz prosta edycja bez maskowania. Wyjątkowa możliwość edycji konwersacyjnej w wielu turach. |
| Dostępność | Ogólnie dostępne | Wersja przedpremierowa (dozwolone użycie w wersji produkcyjnej) |
| Czas oczekiwania | Niska. Zoptymalizowany pod kątem wydajności w czasie zbliżonym do rzeczywistego. | Większą Do korzystania z zaawansowanych funkcji potrzeba większej mocy obliczeniowej. |
| Koszt | Opłacalny w przypadku specjalistycznych zadań. 0,02–0,12 USD za obraz | Ceny oparte na tokenach. 30 USD za 1 mln tokenów w przypadku wygenerowanych obrazów (tokenizacja wygenerowanych obrazów: 1290 tokenów na obraz, maksymalnie 1024 x 1024 piksele) |
| Zalecane zadania |
|
|
Jeśli chcesz zacząć generować obrazy za pomocą Imagen, wybierz model Imagen 4. Wybierz Imagen 4 Ultra, jeśli potrzebujesz zaawansowanych funkcji lub najlepszej jakości obrazu (pamiętaj, że ten model może generować tylko 1 obraz naraz).
Co dalej?
- Więcej przykładów i przykładowych fragmentów kodu znajdziesz w przewodniku.
- Aby dowiedzieć się, jak generować filmy za pomocą interfejsu Gemini API, zapoznaj się z przewodnikiem po Veo.
- Więcej informacji o modelach Gemini znajdziesz w artykule Modele Gemini.