Skip to content

Gemma 4 E4B

by Google · gemma-4 family

4B

parameters

text-generation code-generation reasoning multilingual vision

Gemma 4 E4B is Google's efficient 4.5B-effective-parameter multimodal model, ideal for laptops and consumer devices. It delivers a significant quality leap over its predecessor Gemma 3n E4B while maintaining a similar footprint. Supports text and vision inputs with 128K context. At Q4 it needs about 6 GB of VRAM, fitting comfortably on entry-level GPUs and 8 GB Macs.

Quick Start with Ollama

ollama run e4b-it-q4_K_M
Resources Ollama Hugging Face Official Page
Creator Google
Parameters 4B
Architecture transformer-decoder
Context 128K tokens
Released Apr 2, 2026
License Apache 2.0
Ollama gemma4:e4b

Quantization Options

Format File Size VRAM Required Quality Ollama Tag
Q4_K_M rec 9.6 GB 6 GB e4b-it-q4_K_M
Q8_0 12 GB 10 GB e4b-it-q8_0

Compatible Hardware

Q4_K_M requires 6 GB VRAM

Compatible Hardware

HardwareVRAMTypeFitEst. Speed
Mac Studio M4 Ultra 512GB512 GBmacRuns~137 tok/s
Mac Pro M2 Ultra 192GB192 GBmacRuns~133 tok/s
Mac Studio M4 Ultra 192GB192 GBmacRuns~137 tok/s
Mac Studio M4 Max 128GB128 GBmacRuns~91 tok/s
MacBook Pro M4 Max 128GB128 GBmacRuns~91 tok/s
MacBook Pro M5 Max 128GB128 GBmacRuns~91 tok/s
NVIDIA RTX PRO 6000 Blackwell96 GBgpuRuns~320 tok/s
MacBook Pro M3 Max 96GB96 GBmacRuns~67 tok/s
Mac mini M4 Pro 64GB64 GBmacRuns~46 tok/s
Mac Studio M4 Max 64GB64 GBmacRuns~91 tok/s
MacBook Pro M4 Max 64GB64 GBmacRuns~91 tok/s
MacBook Pro M5 Max 64GB64 GBmacRuns~91 tok/s
NVIDIA RTX 6000 Ada Generation48 GBgpuRuns~160 tok/s
NVIDIA RTX A600048 GBgpuRuns~128 tok/s
NVIDIA RTX PRO 5000 Blackwell48 GBgpuRuns~160 tok/s
Mac mini M4 Pro 48GB48 GBmacRuns~46 tok/s
MacBook Pro M3 Max 48GB48 GBmacRuns~67 tok/s
MacBook Pro M4 Max 48GB48 GBmacRuns~91 tok/s
MacBook Pro M4 Pro 48GB48 GBmacRuns~46 tok/s
MacBook Pro M5 Max 48GB48 GBmacRuns~68 tok/s
MacBook Pro M5 Pro 48GB48 GBmacRuns~46 tok/s
Mac Studio M4 Max 36GB36 GBmacRuns~91 tok/s
MacBook Pro M3 Pro 36GB36 GBmacRuns~25 tok/s
MacBook Pro M5 Max 36GB36 GBmacRuns~68 tok/s
NVIDIA RTX 5000 Ada Generation32 GBgpuRuns~120 tok/s
NVIDIA GeForce RTX 509032 GBgpuRuns~299 tok/s
iMac M4 32GB32 GBmacRuns~20 tok/s
Mac mini M4 32GB32 GBmacRuns~20 tok/s
MacBook Air M4 32GB32 GBmacRuns~20 tok/s
MacBook Air M5 32GB32 GBmacRuns~20 tok/s
MacBook Pro M5 32GB32 GBmacRuns~20 tok/s
AMD Radeon RX 7900 XTX24 GBgpuRuns~160 tok/s
NVIDIA GeForce RTX 309024 GBgpuRuns~156 tok/s
NVIDIA GeForce RTX 3090 Ti24 GBgpuRuns~168 tok/s
NVIDIA GeForce RTX 409024 GBgpuRuns~168 tok/s
NVIDIA RTX A500024 GBgpuRuns~128 tok/s
iMac M3 24GB24 GBmacRuns~17 tok/s
Mac mini M2 24GB24 GBmacRuns~17 tok/s
Mac mini M4 Pro 24GB24 GBmacRuns~46 tok/s
MacBook Air M2 24GB24 GBmacRuns~17 tok/s
MacBook Air M4 24GB24 GBmacRuns~20 tok/s
MacBook Air M5 24GB24 GBmacRuns~20 tok/s
MacBook Pro M4 Pro 24GB24 GBmacRuns~46 tok/s
MacBook Pro M5 24GB24 GBmacRuns~20 tok/s
MacBook Pro M5 Pro 24GB24 GBmacRuns~46 tok/s
AMD Radeon RX 7900 XT20 GBgpuRuns~133 tok/s
NVIDIA RTX 4000 Ada Generation20 GBgpuRuns~60 tok/s
MacBook Pro M3 Pro 18GB18 GBmacRuns~25 tok/s
AMD Radeon RX 6900 XT16 GBgpuRuns~85 tok/s
AMD Radeon RX 6800 XT16 GBgpuRuns~85 tok/s
AMD Radeon RX 7800 XT16 GBgpuRuns~104 tok/s
AMD Radeon RX 9060 XT 16GB16 GBgpuRuns~90 tok/s
AMD Radeon RX 9070 XT16 GBgpuRuns~108 tok/s
AMD Radeon RX 907016 GBgpuRuns~90 tok/s
Intel Arc A77016 GBgpuRuns~93 tok/s
NVIDIA GeForce RTX 4060 Ti 16GB16 GBgpuRuns~48 tok/s
NVIDIA GeForce RTX 4070 Ti Super16 GBgpuRuns~112 tok/s
NVIDIA GeForce RTX 4080 Super16 GBgpuRuns~123 tok/s
NVIDIA GeForce RTX 408016 GBgpuRuns~120 tok/s
NVIDIA GeForce RTX 5060 Ti 16GB16 GBgpuRuns~75 tok/s
NVIDIA GeForce RTX 5070 Ti16 GBgpuRuns~149 tok/s
NVIDIA GeForce RTX 508016 GBgpuRuns~160 tok/s
NVIDIA RTX A400016 GBgpuRuns~75 tok/s
iMac M1 16GB16 GBmacRuns~11 tok/s
iMac M4 16GB16 GBmacRuns~20 tok/s
Mac mini M1 16GB16 GBmacRuns~11 tok/s
Mac mini M4 16GB16 GBmacRuns~20 tok/s
MacBook Air M2 16GB16 GBmacRuns~17 tok/s
MacBook Air M3 16GB16 GBmacRuns~17 tok/s
MacBook Air M4 16GB16 GBmacRuns~20 tok/s
MacBook Air M5 16GB16 GBmacRuns~20 tok/s
MacBook Pro M1 16GB16 GBmacRuns~11 tok/s
MacBook Pro M2 Pro 16GB16 GBmacRuns~33 tok/s
MacBook Pro M5 16GB16 GBmacRuns~20 tok/s
AMD Radeon RX 6700 XT12 GBgpuRuns~64 tok/s
AMD Radeon RX 7700 XT12 GBgpuRuns~72 tok/s
Intel Arc B58012 GBgpuRuns~76 tok/s
NVIDIA GeForce RTX 3060 12GB12 GBgpuRuns~60 tok/s
NVIDIA GeForce RTX 3080 12GB12 GBgpuRuns~152 tok/s
NVIDIA GeForce RTX 4070 Super12 GBgpuRuns~84 tok/s
NVIDIA GeForce RTX 4070 Ti12 GBgpuRuns~84 tok/s
NVIDIA GeForce RTX 407012 GBgpuRuns~84 tok/s
NVIDIA GeForce RTX 507012 GBgpuRuns~112 tok/s
NVIDIA GeForce GTX 1080 Ti11 GBgpuRuns~81 tok/s
NVIDIA GeForce RTX 2080 Ti11 GBgpuRuns~103 tok/s
Intel Arc B57010 GBgpuRuns~63 tok/s
NVIDIA GeForce RTX 3080 10GB10 GBgpuRuns~127 tok/s
AMD Radeon RX 6600 XT8 GBgpuRuns~43 tok/s
AMD Radeon RX 76008 GBgpuRuns~48 tok/s
AMD Radeon RX 9060 XT 8GB8 GBgpuRuns~45 tok/s
Intel Arc A7508 GBgpuRuns~85 tok/s
NVIDIA GeForce GTX 10708 GBgpuRuns~43 tok/s
NVIDIA GeForce RTX 2060 Super8 GBgpuRuns~75 tok/s
NVIDIA GeForce RTX 2070 Super8 GBgpuRuns~75 tok/s
NVIDIA GeForce RTX 2080 Super8 GBgpuRuns~83 tok/s
NVIDIA GeForce RTX 30508 GBgpuRuns~37 tok/s
NVIDIA GeForce RTX 30708 GBgpuRuns~75 tok/s
NVIDIA GeForce RTX 3060 Ti8 GBgpuRuns~75 tok/s
NVIDIA GeForce RTX 4060 Ti 8GB8 GBgpuRuns~48 tok/s
NVIDIA GeForce RTX 40608 GBgpuRuns~45 tok/s
NVIDIA GeForce RTX 50508 GBgpuRuns~37 tok/s
NVIDIA GeForce RTX 5060 Ti 8GB8 GBgpuRuns~75 tok/s
NVIDIA GeForce RTX 50608 GBgpuRuns~56 tok/s
MacBook Air M1 8GB8 GBmacRuns~11 tok/s
MacBook Air M2 8GB8 GBmacRuns~17 tok/s
NVIDIA GeForce GTX 1660 Super6 GBgpuCPU Offload~17 tok/s
NVIDIA GeForce RTX 20606 GBgpuCPU Offload~17 tok/s