Gemma 4 E2B
google/gemma-4-e2b-it
Parameters
5.1B
Context
128K
Downloads
3.8M
Likes
883
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 17.0 GB RAM 32.0 · Disk 53.8 | 33.3 GB RAM 49.9 · Disk 79.8 | 130.7 GB RAM 196.1 · Disk 118.8 |
| FP88bit | 11.0 GB RAM 32.0 · Disk 46.4 | 27.0 GB RAM 40.5 · Disk 72.4 | 124.2 GB RAM 186.2 · Disk 111.4 |
| INT88bit | 11.0 GB RAM 32.0 · Disk 46.4 | 27.0 GB RAM 40.5 · Disk 72.4 | 124.2 GB RAM 186.2 · Disk 111.4 |
| AWQ4bit | 8.1 GB RAM 32.0 · Disk 42.8 | 24.0 GB RAM 35.9 · Disk 68.8 | 121.0 GB RAM 181.5 · Disk 107.8 |
| GPTQ4bit | 8.1 GB RAM 32.0 · Disk 42.9 | 24.0 GB RAM 36.0 · Disk 68.9 | 121.0 GB RAM 181.6 · Disk 107.9 |
| GGUF4bit | 8.2 GB RAM 32.0 · Disk 43.0 | 24.1 GB RAM 36.1 · Disk 69.0 | 121.1 GB RAM 181.7 · Disk 108.0 |
GPU Recommendations
Best ValueBEST
1× RTX 5090
32 GB VRAM
$313/mo
≈ 728.5 tok/s
Effective production throughput ≈ 510 tok/s
score 0.6959
Cheapest
1× L4
24 GB VRAM
$76.2/mo
≈ 122 tok/s
Effective production throughput ≈ 85.4 tok/s
score 0.6372
Performance
8× H200 SXM
1128 GB VRAM
$20,440/mo
≈ 6243.9 tok/s
Effective production throughput ≈ 4370.7 tok/s
score 0.2197
Min Complexity
1× RTX 3090
24 GB VRAM
$88.4/mo
≈ 380.5 tok/s
Effective production throughput ≈ 266.3 tok/s
score 0.6415
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
L4
24 GB VRAM
$76.2
/mo · vast
RTX 3090
24 GB VRAM
$88.4
/mo · vast
RTX 4090
24 GB VRAM
$190
/mo · tensordock
A100 SXM4
80 GB VRAM
$225
/mo · vast
RTX A6000
48 GB VRAM
$294
/mo · vast