back

Gemma 4 E4B IT

Gemma

Google · 8B · Dense

Gemma 4 balanced instruct model (official)

590.5K downloads 62 likes 2026-04 256K context

Use Cases

chat vision

Quantization Options

Quant Bits VRAM Quality Status
Q4_K_M 4 4.6 GB good —
Q6_K 6 6.6 GB excellent —
Q8_0 8 8.7 GB excellent —

GPUs that run Gemma 4 E4B IT

Gemma 4 E4B IT needs about 4.6 GB of VRAM at Q4_K_M, so it runs comfortably on a GPU with 6 GB of VRAM or more (157 of the GPUs we track). Common cards, smallest first:

About this model

Gemma 4 balanced instruct model (official)