Gemma2 2B
google/gemma-2-2b-it
Parameters
2.6B
Context
8K
Downloads
670K
Likes
1.4K
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 11.1 GB RAM 32.0 · Disk 46.6 | 27.1 GB RAM 40.7 · Disk 72.6 | 124.3 GB RAM 186.4 · Disk 111.6 |
| FP88bit | 8.0 GB RAM 32.0 · Disk 42.8 | 23.9 GB RAM 35.9 · Disk 68.8 | 120.9 GB RAM 181.4 · Disk 107.8 |
| INT88bit | 8.0 GB RAM 32.0 · Disk 42.8 | 23.9 GB RAM 35.9 · Disk 68.8 | 120.9 GB RAM 181.4 · Disk 107.8 |
| AWQ4bit | 6.5 GB RAM 32.0 · Disk 41.0 | 22.4 GB RAM 33.5 · Disk 67.0 | 119.3 GB RAM 179.0 · Disk 106.0 |
| GPTQ4bit | 6.6 GB RAM 32.0 · Disk 41.0 | 22.4 GB RAM 33.6 · Disk 67.0 | 119.4 GB RAM 179.0 · Disk 106.0 |
| GGUF4bit | 6.6 GB RAM 32.0 · Disk 41.0 | 22.4 GB RAM 33.6 · Disk 67.0 | 119.4 GB RAM 179.1 · Disk 106.0 |
GPU Recommendations
Best ValueBEST
1× RTX 5090
32 GB VRAM
$343/mo
≈ 1422.2 tok/s
Effective production throughput ≈ 995.5 tok/s
score 0.682
Cheapest
1× RTX 3090
24 GB VRAM
$89.2/mo
≈ 742.9 tok/s
Effective production throughput ≈ 520 tok/s
score 0.6464
Performance
8× H200 SXM
1128 GB VRAM
$20,440/mo
≈ 12190.5 tok/s
Effective production throughput ≈ 8533.4 tok/s
score 0.2197
Min Complexity
1× RTX 4090
24 GB VRAM
$190/mo
≈ 800 tok/s
Effective production throughput ≈ 560 tok/s
score 0.6448
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX 3090
24 GB VRAM
$89.2
/mo · vast
A10
24 GB VRAM
$171
/mo · vast
L4
24 GB VRAM
$185
/mo · vast
RTX 4090
24 GB VRAM
$190
/mo · tensordock
RTX A6000
48 GB VRAM
$210
/mo · vast