# Not able to run google/gemma-3n-E4B-it **URL:** \ **Category:** General\ **Created:** [September 22, 2025, 9:23pm UTC](https://discuss.vllm.ai/t/not-able-to-run-google-gemma-3n-e4b-it/1658 "2025-09-22T21:23:10Z")\ **Posts on this page:** 4\ **Page:** 1
**Author:** ![youlearnit](https://avatars.discourse-cdn.com/v4/letter/y/71c47a/32.png) [@youlearnit](https://discuss.vllm.ai/u/youlearnit)\ **Post date:** [September 22, 2025, 9:23pm UTC](https://discuss.vllm.ai/t/not-able-to-run-google-gemma-3n-e4b-it/1658/1 "2025-09-22T21:23:10Z")
Not able to run google/gemma-3n-E4B-it on local vllm with 2 gpus. Trying to run it with: docker run -e VLLM\_SLEEP\_WHEN\_IDLE=1 -it --gpus all --ipc=host --network host -p 8000:8000 -v /home/ubuntu/vllm\_models:/root/.cache/huggingface -e HUGGING\_FACE\_HUB\_TOKEN=“” --name vllm\_tp2-gemma-3n-E4B-it vllm-openai-nccl --tensor-parallel-size 2 --model google/gemma-3n-E4B-it --served-model-name vllm/gemma-3n --max-model-len 2048 --max-num-seqs 64 gemma-3-12b-it works fine. ImportError: The model architecture ‘mobilenetv5\_300m\_enc’ is not supported in your version of timm (0.9.10). Please upgrade timm to a more recent version with `pip install -U timm`. (VllmWorker TP0 pid=152) INFO 09-22 14:18:21 [multiproc\_executor.py:520] Parent process exited, terminating worker (VllmWorker TP1 pid=153) INFO 09-22 14:18:21 [multiproc\_executor.py:520] Parent process exited, terminating worker [rank0]:[W922 14:18:22.767945607 ProcessGroupNCCL.cpp:1479] Warning: WARNING: destroy\_process\_group() was not called before program exit, which can leak resources. For more info, please see [Distributed communication package - torch.distributed — PyTorch 2.8 documentation](https://pytorch.org/docs/stable/distributed.html#shutdown) (function operator()) ---
**Author:** ![RunLLM](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.vllm.ai/runllm/32/233_2.png) [@RunLLM](https://discuss.vllm.ai/u/RunLLM)\ **Post date:** [September 22, 2025, 9:23pm UTC](https://discuss.vllm.ai/t/not-able-to-run-google-gemma-3n-e4b-it/1658/2 "2025-09-22T21:23:32Z")
The error occurs because the version of `timm` in your Docker image (0.9.10) does not support the `mobilenetv5_300m_enc` architecture required by Gemma 3n models. You need to upgrade `timm` to at least 1.0.17 and also ensure `transformers` is at least 4.55.4 for full Gemma 3n support. This is a known issue and has been fixed in newer Docker images and code; as a workaround, you can run `pip install --upgrade timm==1.0.19 transformers==4.55.4` inside your container before starting vLLM, or rebuild your Docker image with these versions pre-installed. See the official fix and workaround details in [this issue](https://github.com/vllm-project/vllm/issues/23821) and [this PR](https://github.com/vllm-project/vllm/pull/23857). Would you like step-by-step instructions for updating your Docker image or container? Sources: - [https://github.com/vllm-project/vllm/issues/23821](https://github.com/vllm-project/vllm/issues/23821) - [https://github.com/vllm-project/vllm/pull/23857](https://github.com/vllm-project/vllm/pull/23857) ---
**Author:** ![youlearnit](https://avatars.discourse-cdn.com/v4/letter/y/71c47a/32.png) [@youlearnit](https://discuss.vllm.ai/u/youlearnit)\ **Post date:** [September 22, 2025, 9:26pm UTC](https://discuss.vllm.ai/t/not-able-to-run-google-gemma-3n-e4b-it/1658/3 "2025-09-22T21:26:03Z")
how can I get the latest docker image where these are fixed? sudo docker pull vllm/vllm-openai:latest have done that but looks like its not the latest? ---
**Author:** ![youlearnit](https://avatars.discourse-cdn.com/v4/letter/y/71c47a/32.png) [@youlearnit](https://discuss.vllm.ai/u/youlearnit)\ **Post date:** [September 22, 2025, 10:08pm UTC](https://discuss.vllm.ai/t/not-able-to-run-google-gemma-3n-e4b-it/1658/4 "2025-09-22T22:08:14Z")
upgraded the timm, but now getting this: ERROR 09-22 15:07:24 [core.py:700] RuntimeError: Worker failed with error ‘mat1 and mat2 shapes cannot be multiplied (1792x2048 and 1024x2048)’, please check the stack trace above for the root cause (EngineCore\_0 pid=102) ERROR 09-22 15:07:26 [multiproc\_executor.py:146] Worker proc VllmWorker-1 died unexpectedly, shutting down executor. (EngineCore\_0 pid=102) Process EngineCore\_0: (EngineCore\_0 pid=102) Traceback (most recent call last): (EngineCore\_0 pid=102) File “/usr/lib/python3.12/multiprocessing/process.py”, line 314, in \_bootstrap (EngineCore\_0 pid=102) self.run() (EngineCore\_0 pid=102) File “/usr/lib/python3.12/multiprocessing/process.py”, line 108, in run (EngineCore\_0 pid=102) self.\_target(\*self.\_args, \*\*self.\_kwargs) (EngineCore\_0 pid=102) File “/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core.py”, line 704, in run\_engine\_core (EngineCore\_0 pid=102) raise e (EngineCore\_0 pid=102) File “/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core.py”, line 691, in run\_engine\_core (EngineCore\_0 pid=102) engine\_core = EngineCoreProc(\*args, \*\*kwargs) (EngineCore\_0 pid=102) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ (EngineCore\_0 pid=102) File “/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core.py”, line 492, in **init** (EngineCore\_0 pid=102) super(). **init** (vllm\_config, executor\_class, log\_stats, (EngineCore\_0 pid=102) File “/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core.py”, line 89, in **init** (EngineCore\_0 pid=102) self.\_initialize\_kv\_caches(vllm\_config) (EngineCore\_0 pid=102) File “/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core.py”, line 179, in \_initialize\_kv\_caches (EngineCore\_0 pid=102) self.model\_executor.determine\_available\_memory()) (EngineCore\_0 pid=102) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ (EngineCore\_0 pid=102) File “/usr/local/lib/python3.12/dist-packages/vllm/v1/executor/abstract.py”, line 76, in determine\_available\_memory (EngineCore\_0 pid=102) output = self.collective\_rpc(“determine\_available\_memory”) (EngineCore\_0 pid=102) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ (EngineCore\_0 pid=102) File “/usr/local/lib/python3.12/dist-packages/vllm/v1/executor/multiproc\_executor.py”, line 243, in collective\_rpc (EngineCore\_0 pid=102) result = get\_response(w, dequeue\_timeout) (EngineCore\_0 pid=102) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ (EngineCore\_0 pid=102) File “/usr/local/lib/python3.12/dist-packages/vllm/v1/executor/multiproc\_executor.py”, line 230, in get\_response (EngineCore\_0 pid=102) raise RuntimeError( (EngineCore\_0 pid=102) RuntimeError: Worker failed with error ‘mat1 and mat2 shapes cannot be multiplied (1792x2048 and 1024x2048)’, please check the stack trace above for the root cause (APIServer pid=1) Traceback (most recent call last):