community pool / model answer

gemma-4-E4B-it-GGUF

chat · ggml-org/gemma-4-E4B-it-GGUF

95.3 tok/s · model median · 5 runs in the model record.

Reference rigQuantizationMedian resultBasisTTFT (ms)Peak VRAMMax contextEngines
RTX A5000 24GB4-bit121 tok/sreported-3.7 GB4,096llama.cpp
RTX A5000 24GB8-bit95.3 tok/sreported-5.6 GB4,096llama.cpp
TECNO POVA 7 Ultra 5G / Mali-G615 MC6 11.23GB4-bit7.1 tok/sreported-- GB640llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.