Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Appearance settings

Commit eb074de

Browse filesBrowse files
msampathkumarmsampathkumar
authored
feat(image-generation): Image generation using Gemini Flash (GoogleCloudPlatform#13334)
* feat(image-generation): Image generation using Gemini Flash * fix(image-generation): update model name * chore: code refactor * fix: reducing tests to only 3.12 due to frequent CICD `RESOURCE_EXHAUSTED` error
1 parent 062a5af commit eb074de
Copy full SHA for eb074de

File tree

Expand file treeCollapse file tree

5 files changed

+189
-10
lines changed
Filter options
Expand file treeCollapse file tree

5 files changed

+189
-10
lines changed

‎genai/image_generation/imggen_mmflash_edit_img_with_txt_img.py

Copy file name to clipboard
+50Lines changed: 50 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,50 @@
1+
# Copyright 2025 Google LLC
2+
#
3+
# Licensed under the Apache License, Version 2.0 (the "License");
4+
# you may not use this file except in compliance with the License.
5+
# You may obtain a copy of the License at
6+
#
7+
# https://www.apache.org/licenses/LICENSE-2.0
8+
#
9+
# Unless required by applicable law or agreed to in writing, software
10+
# distributed under the License is distributed on an "AS IS" BASIS,
11+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12+
# See the License for the specific language governing permissions and
13+
# limitations under the License.
14+
15+
16+
def generate_content() -> str:
17+
# [START googlegenaisdk_imggen_mmflash_edit_img_with_txt_img]
18+
from google import genai
19+
from google.genai.types import GenerateContentConfig, Modality
20+
from PIL import Image
21+
from io import BytesIO
22+
23+
client = genai.Client()
24+
25+
# Using an image of Eiffel tower, with fireworks in the background.
26+
image = Image.open("example-image.png")
27+
28+
response = client.models.generate_content(
29+
model="gemini-2.0-flash-exp",
30+
contents=[image, "Edit this image to make it look like a cartoon."],
31+
config=GenerateContentConfig(response_modalities=[Modality.TEXT, Modality.IMAGE]),
32+
)
33+
for part in response.candidates[0].content.parts:
34+
if part.text:
35+
print(part.text)
36+
elif part.inline_data:
37+
image = Image.open(BytesIO((part.inline_data.data)))
38+
image.save("bw-example-image.png")
39+
# Example response:
40+
# Here's the cartoon-style edit of the image:
41+
# Cartoon-style edit:
42+
# - Simplified the Eiffel Tower with bolder lines and slightly exaggerated proportions.
43+
# - Brightened and saturated the colors of the sky, fireworks, and foliage for a more vibrant, cartoonish look.
44+
# ....
45+
# [END googlegenaisdk_imggen_mmflash_edit_img_with_txt_img]
46+
return "bw-example-image.png"
47+
48+
49+
if __name__ == "__main__":
50+
generate_content()

‎genai/image_generation/imggen_mmflash_txt_and_img_with_txt.py

Copy file name to clipboard
+49Lines changed: 49 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,49 @@
1+
# Copyright 2025 Google LLC
2+
#
3+
# Licensed under the Apache License, Version 2.0 (the "License");
4+
# you may not use this file except in compliance with the License.
5+
# You may obtain a copy of the License at
6+
#
7+
# https://www.apache.org/licenses/LICENSE-2.0
8+
#
9+
# Unless required by applicable law or agreed to in writing, software
10+
# distributed under the License is distributed on an "AS IS" BASIS,
11+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12+
# See the License for the specific language governing permissions and
13+
# limitations under the License.
14+
15+
16+
def generate_content() -> int:
17+
# [START googlegenaisdk_imggen_mmflash_txt_and_img_with_txt]
18+
from google import genai
19+
from google.genai.types import GenerateContentConfig, Modality
20+
from PIL import Image
21+
from io import BytesIO
22+
23+
client = genai.Client()
24+
25+
response = client.models.generate_content(
26+
model="gemini-2.0-flash-exp",
27+
contents=(
28+
"Generate an illustrated recipe for a paella."
29+
"Create images to go alongside the text as you generate the recipe"
30+
),
31+
config=GenerateContentConfig(response_modalities=[Modality.TEXT, Modality.IMAGE]),
32+
)
33+
with open("paella-recipe.md", "w") as fp:
34+
for i, part in enumerate(response.candidates[0].content.parts):
35+
if part.text is not None:
36+
fp.write(part.text)
37+
elif part.inline_data is not None:
38+
image = Image.open(BytesIO((part.inline_data.data)))
39+
image.save(f"example-image-{i+1}.png")
40+
fp.write(f"![image](./example-image-{i+1}.png)")
41+
# Example response:
42+
# A markdown page for a Paella recipe(`paella-recipe.md`) has been generated.
43+
# It includes detailed steps and several images illustrating the cooking process.
44+
# [END googlegenaisdk_imggen_mmflash_txt_and_img_with_txt]
45+
return i
46+
47+
48+
if __name__ == "__main__":
49+
generate_content()

‎genai/image_generation/imggen_mmflash_with_txt.py

Copy file name to clipboard
+46Lines changed: 46 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,46 @@
1+
# Copyright 2025 Google LLC
2+
#
3+
# Licensed under the Apache License, Version 2.0 (the "License");
4+
# you may not use this file except in compliance with the License.
5+
# You may obtain a copy of the License at
6+
#
7+
# https://www.apache.org/licenses/LICENSE-2.0
8+
#
9+
# Unless required by applicable law or agreed to in writing, software
10+
# distributed under the License is distributed on an "AS IS" BASIS,
11+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12+
# See the License for the specific language governing permissions and
13+
# limitations under the License.
14+
15+
16+
def generate_content() -> str:
17+
# [START googlegenaisdk_imggen_mmflash_with_txt]
18+
from google import genai
19+
from google.genai.types import GenerateContentConfig, Modality
20+
from PIL import Image
21+
from io import BytesIO
22+
23+
client = genai.Client()
24+
25+
response = client.models.generate_content(
26+
model="gemini-2.0-flash-exp",
27+
contents=(
28+
"Generate an image of the Eiffel tower with fireworks in the background."
29+
),
30+
config=GenerateContentConfig(response_modalities=[Modality.TEXT, Modality.IMAGE]),
31+
)
32+
for part in response.candidates[0].content.parts:
33+
if part.text:
34+
print(part.text)
35+
elif part.inline_data:
36+
image = Image.open(BytesIO((part.inline_data.data)))
37+
image.save("example-image.png")
38+
# Example response:
39+
# A beautiful photograph captures the iconic Eiffel Tower in Paris, France,
40+
# against a backdrop of a vibrant and dynamic fireworks display. The tower itself...
41+
# [END googlegenaisdk_imggen_mmflash_with_txt]
42+
return "example-image.png"
43+
44+
45+
if __name__ == "__main__":
46+
generate_content()

‎genai/image_generation/noxfile_config.py

Copy file name to clipboardExpand all lines: genai/image_generation/noxfile_config.py
+1-1Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -22,7 +22,7 @@
2222

2323
TEST_CONFIG_OVERRIDE = {
2424
# You can opt out from the test for specific Python versions.
25-
"ignored_versions": ["2.7", "3.7", "3.8", "3.10", "3.11", "3.13"],
25+
"ignored_versions": ["2.7", "3.7", "3.8", "3.9", "3.10", "3.11", "3.13"],
2626
# Old samples are opted out of enforcing Python type hints
2727
# All new samples should feature them
2828
"enforce_type_hints": True,

‎genai/image_generation/test_image_generation.py

Copy file name to clipboardExpand all lines: genai/image_generation/test_image_generation.py
+43-9Lines changed: 43 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -25,13 +25,12 @@
2525
import pytest
2626

2727
import imggen_canny_ctrl_type_with_txt_img
28-
28+
import imggen_mmflash_edit_img_with_txt_img
29+
import imggen_mmflash_txt_and_img_with_txt
30+
import imggen_mmflash_with_txt
2931
import imggen_raw_reference_with_txt_img
30-
3132
import imggen_scribble_ctrl_type_with_txt_img
32-
3333
import imggen_style_reference_with_txt_img
34-
3534
import imggen_subj_refer_ctrl_refer_with_txt_imgs
3635

3736

@@ -57,25 +56,60 @@ def output_gcs_uri() -> str:
5756

5857

5958
def test_img_customization_subject(output_gcs_uri: str) -> None:
60-
response = imggen_subj_refer_ctrl_refer_with_txt_imgs.subject_customization(output_gcs_uri=output_gcs_uri)
59+
response = imggen_subj_refer_ctrl_refer_with_txt_imgs.subject_customization(
60+
output_gcs_uri=output_gcs_uri
61+
)
6162
assert response
6263

6364

6465
def test_img_customization_style(output_gcs_uri: str) -> None:
65-
response = imggen_style_reference_with_txt_img.style_customization(output_gcs_uri=output_gcs_uri)
66+
response = imggen_style_reference_with_txt_img.style_customization(
67+
output_gcs_uri=output_gcs_uri
68+
)
6669
assert response
6770

6871

6972
def test_img_customization_style_transfer(output_gcs_uri: str) -> None:
70-
response = imggen_raw_reference_with_txt_img.style_transfer_customization(output_gcs_uri=output_gcs_uri)
73+
response = imggen_raw_reference_with_txt_img.style_transfer_customization(
74+
output_gcs_uri=output_gcs_uri
75+
)
7176
assert response
7277

7378

7479
def test_img_customization_scribble(output_gcs_uri: str) -> None:
75-
response = imggen_scribble_ctrl_type_with_txt_img.scribble_customization(output_gcs_uri=output_gcs_uri)
80+
response = imggen_scribble_ctrl_type_with_txt_img.scribble_customization(
81+
output_gcs_uri=output_gcs_uri
82+
)
7683
assert response
7784

7885

7986
def test_img_customization_canny_edge(output_gcs_uri: str) -> None:
80-
response = imggen_canny_ctrl_type_with_txt_img.canny_edge_customization(output_gcs_uri=output_gcs_uri)
87+
response = imggen_canny_ctrl_type_with_txt_img.canny_edge_customization(
88+
output_gcs_uri=output_gcs_uri
89+
)
8190
assert response
91+
92+
93+
def test_imggen_mmflash_examples() -> None:
94+
# generate image
95+
fname = imggen_mmflash_with_txt.generate_content()
96+
assert os.path.isfile(fname)
97+
# edit generate image
98+
new_fname = imggen_mmflash_edit_img_with_txt_img.generate_content()
99+
assert os.path.isfile(new_fname)
100+
101+
# clean-up
102+
os.remove(fname)
103+
os.remove(new_fname)
104+
105+
106+
def test_imggen_mmflash_txt_and_img_with_txt() -> None:
107+
last_image_id = imggen_mmflash_txt_and_img_with_txt.generate_content()
108+
# clean-up
109+
for i in range(last_image_id + 1):
110+
img_name = f"example-image-{i+1}.png"
111+
if os.path.isfile(img_name):
112+
os.remove(img_name)
113+
fname = "paella-recipe.md"
114+
if os.path.isfile(fname):
115+
os.remove(fname)

0 commit comments

Comments
0 (0)
Morty Proxy This is a proxified and sanitized view of the page, visit original site.