facebook/dinov2-small
facebook/dinov2-small
Parameters
33.5M
Context
—
Downloads
4.3M
Likes
70
Architecture
dinov2
12 layers · head 64 · hidden 384
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 2.5 GB RAM 32.0 · Disk 39.1 | 4.4 GB RAM 32.0 · Disk 65.1 | 18.2 GB RAM 32.0 · Disk 104.1 |
| FP88bit | 2.4 GB RAM 32.0 · Disk 39.1 | 4.4 GB RAM 32.0 · Disk 65.1 | 18.2 GB RAM 32.0 · Disk 104.1 |
| INT88bit | 2.4 GB RAM 32.0 · Disk 39.1 | 4.4 GB RAM 32.0 · Disk 65.1 | 18.2 GB RAM 32.0 · Disk 104.1 |
| AWQ4bit | 2.4 GB RAM 32.0 · Disk 39.0 | 4.4 GB RAM 32.0 · Disk 65.0 | 18.2 GB RAM 32.0 · Disk 104.0 |
| GPTQ4bit | 2.4 GB RAM 32.0 · Disk 39.0 | 4.4 GB RAM 32.0 · Disk 65.0 | 18.2 GB RAM 32.0 · Disk 104.0 |
| GGUF4bit | 2.4 GB RAM 32.0 · Disk 39.0 | 4.4 GB RAM 32.0 · Disk 65.0 | 18.2 GB RAM 32.0 · Disk 104.0 |
GPU Recommendations
1× RTX 5080
16 GB VRAM
≈ 48000 tok/s
Effective production throughput ≈ 33600 tok/s
score 0.5678
1× RTX 5060 Ti
8 GB VRAM
≈ 22400 tok/s
Effective production throughput ≈ 15680 tok/s
score 0.638
8× H200 SXM
1128 GB VRAM
≈ 768000 tok/s
Effective production throughput ≈ 537600 tok/s
score 0.2197
1× RTX 5070
12 GB VRAM
≈ 33600 tok/s
Effective production throughput ≈ 23520 tok/s
score 0.5908
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX 5060 Ti
8 GB VRAM
$39.7
/mo · vast
RTX 4070S Ti
16 GB VRAM
$49.5
/mo · vast
RTX 4080S
16 GB VRAM
$50.0
/mo · vast
RTX 5070
12 GB VRAM
$54.1
/mo · vast
RTX 5070 Ti
16 GB VRAM
$83.1
/mo · vast