This is a community-quantized version of the Gemma 4 model using E4B (4-bit embedding) weights and 16-bit activations. It is instruction-tuned and includes a specialized chat template. The model is hosted on Hugging Face and intended for efficient inference in text generation applications.
Gemma 4 E4B It W4A16 sits in PulseGate's Foundation models & chat category. It focuses on running large Gemma models with reduced memory footprint through 4-bit quantization. It is built as an open-source project for machine learning developers. Gemma 4 E4B It W4A16 is open source under the Open Source license. Gemma 4 E4B It W4A16 is available on the web, the command line, and API.
It is developed by ciocan, and the product first shipped in 2025. The category is crowded — PulseGate's index counts 25 comparable apps. Key capabilities include Quantized Weights, Instruction Tuning, and Chat Template.
Latest indexed changes and source events
ciocan/gemma-4-E4B-it-W4A16 verified by the PulseGate indexer
Other apps tracked under the same category.
