Skip to content
Log in

google/gemma-4-E4B-it

huggingface_fp16_tensorrt-edgellm
google/gemma-4-E4B-it architecture graph
google/gemma-4-E4B-it architecture graph
  • huggingface
  • AGX Thor
Labels
  • Image
  • Text
  • VLM
  • Generation
Build
Embedl-optimized
Precision
W4A16
Runtime
TensorRT Edge-LLM 0.10.1 with Embedl optimizations
Context length
4,096 tokens

Google's Gemma 4 model for on-device use, with about 4B effective parameters, with images.

Files

Image Size

Model

docker.io/embedl/model-artifacts:gemma4-e4b.huggingface_int4_embedl-flashhead-mtp_tensorrt-edgellm.a58ffb8a50db7d28

 10.35 GiB

Runtime

docker.io/embedl/models:runtimes.tensorrt-edgellm.arm64.2026101815014ee0

 1.01 GiB