protobuf issue

[aryan-rathore13]

Description of the bug:

I am following this guide https://colab.research.google.com/github/google-gemini/gemma-cookbook/blob/main/Demos/Emoji-Gemma-on-Web/resources/Convert_Gemma_3_270M_to_LiteRT_for_MediaPipe_LLM_Inference_API.ipynb to convert my fine tuned gemma 270M model into a task file. When I'm trying to install the dependencies , I am getting this error in colab:
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed.
grpcio-status 1.71.2 requires protobuf<6.0dev,>=5.26.1, but you have protobuf 6.32.1.
google-ai-generativelanguage 0.6.15 requires protobuf<6.0.0dev,>=3.20.2, but you have protobuf 6.32.1.
Pip resolves to protobuf 6.32.1 (required by tf-nightly 2.21.0.dev20250819 and ai-edge-tensorflow 2.21.0.dev20250818), but several Google packages in the same environment still require protobuf < 6.0.
Can anyone guide me how to solve this? or if there is any other guide which I can follow to convert to .tflite and hence .task file for my fine tuned model.

Actual vs expected behavior:

No response

Any other information you'd like to share?

No response

[aryan-rathore13]

https://ai.google.dev/gemma/docs/conversions/hf-to-mediapipe-task I have tried this guide as well and I get the following issue.
My code:
from ai_edge_torch.generative.examples.gemma3 import gemma3
from ai_edge_torch.generative.utilities import converter
from ai_edge_torch.generative.utilities.export_config import ExportConfig
from ai_edge_torch.generative.layers import kv_cache

pytorch_model = gemma3.build_model_270m("content/drive/MyDrive/ShakespeareGemma")

export_config = ExportConfig()
export_config.kvcache_layout = kv_cache.KV_LAYOUT_TRANSPOSED
export_config.mask_as_input = True

converter.convert_to_tflite(
pytorch_model,
output_path="content/drive/MyDrive/tflite_file",
output_name_prefix="my-gemma3",
prefill_seq_len=2048,
kv_cache_max_len=4096,
quantize="dynamic_int8",
export_config=export_config,
)
Error : NotFoundError Traceback (most recent call last)
/tmp/ipython-input-3487267011.py in <cell line: 0>()
----> 1 from ai_edge_torch.generative.examples.gemma3 import gemma3
2 from ai_edge_torch.generative.utilities import converter
3 from ai_edge_torch.generative.utilities.export_config import ExportConfig
4 from ai_edge_torch.generative.layers import kv_cache
5

4 frames
/usr/local/lib/python3.12/dist-packages/tensorflow/python/framework/load_library.py in load_library(library_location)
149
150 for lib in kernel_libraries:
--> 151 py_tf.TF_LoadLibrary(lib)
152
153 else:

NotFoundError: /usr/local/lib/python3.12/dist-packages/tensorflow/core/kernels/libtfkernel_sobol_op.so: undefined symbol: _ZN10tensorflow15TensorShapeBaseINS_11TensorShapeEEC2EN4absl12lts_202308024SpanIKlEE

[gaikwadrahul8]

Hi, @aryan-rathore13
Thank you for your patience and for reporting this issue. I apologize for the delayed response. I have run the notebook as provided without any modifications and was unable to reproduce the error. The notebook completed execution successfully on my end. For your reference, here is a gist-file of my execution

Could you please try running the notebook again and let me know if you are still encountering the same issue? Thank you for your cooperation.

Here is output screenshot for reference :

[github-actions]

Marking this issue as stale since it has been open for 7 days with no activity. This issue will be closed if no further activity occurs.

[github-actions]

This issue was closed because it has been inactive for 14 days. Please post a new issue if you need further assistance. Thanks!