InternVL2 26B
OpenGVLab/InternVL2-26B
Parameters
25.5B
Context
8K
Downloads
1.1K
Likes
118
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 63.7 GB RAM 95.5 · Disk 113.1 | 68.6 GB RAM 102.9 · Disk 139.1 | 87.2 GB RAM 130.7 · Disk 178.1 |
| FP88bit | 33.6 GB RAM 50.4 · Disk 76.1 | 37.2 GB RAM 55.7 · Disk 102.1 | 54.4 GB RAM 81.6 · Disk 141.1 |
| INT88bit | 33.6 GB RAM 50.4 · Disk 76.1 | 37.2 GB RAM 55.7 · Disk 102.1 | 54.4 GB RAM 81.6 · Disk 141.1 |
| AWQ4bit | 19.2 GB RAM 32.0 · Disk 58.2 | 22.0 GB RAM 33.1 · Disk 84.2 | 38.6 GB RAM 57.9 · Disk 123.2 |
| GPTQ4bit | 19.4 GB RAM 32.0 · Disk 58.5 | 22.2 GB RAM 33.4 · Disk 84.5 | 38.8 GB RAM 58.2 · Disk 123.5 |
| GGUF4bit | 19.7 GB RAM 32.0 · Disk 58.9 | 22.6 GB RAM 34.0 · Disk 84.9 | 39.2 GB RAM 58.8 · Disk 123.9 |
GPU Recommendations
1× RTX 5090
32 GB VRAM
≈ 145.5 tok/s
Effective production throughput ≈ 101.9 tok/s
score 0.6816
2× Tesla V100
32 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.6303
8× H200 SXM
1128 GB VRAM
≈ 1246.8 tok/s
Effective production throughput ≈ 872.8 tok/s
score 0.2197
1× RTX 3090
24 GB VRAM
≈ 76 tok/s
Effective production throughput ≈ 53.2 tok/s
score 0.6501
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX 3090
24 GB VRAM
$79.5
/mo · vast
RTX 4090
24 GB VRAM
$98.1
/mo · vast
RTX 4090D
24 GB VRAM
$117
/mo · vast
L4
24 GB VRAM
$147
/mo · vast
A10
24 GB VRAM
$176
/mo · vast