FacebookAI/roberta-large
FacebookAI/roberta-large
Parameters
353.5M
Context
514
Downloads
11.4M
Likes
319
Architecture
roberta
24 layers · head 64 · hidden 1024
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 3.7 GB RAM 32.0 · Disk 40.0 | 8.4 GB RAM 32.0 · Disk 66.0 | 39.3 GB RAM 59.0 · Disk 105.0 |
| FP88bit | 3.2 GB RAM 32.0 · Disk 39.5 | 7.9 GB RAM 32.0 · Disk 65.5 | 38.9 GB RAM 58.3 · Disk 104.5 |
| INT88bit | 3.2 GB RAM 32.0 · Disk 39.5 | 7.9 GB RAM 32.0 · Disk 65.5 | 38.9 GB RAM 58.3 · Disk 104.5 |
| AWQ4bit | 3.0 GB RAM 32.0 · Disk 39.3 | 7.7 GB RAM 32.0 · Disk 65.3 | 38.6 GB RAM 58.0 · Disk 104.3 |
| GPTQ4bit | 3.0 GB RAM 32.0 · Disk 39.3 | 7.7 GB RAM 32.0 · Disk 65.3 | 38.6 GB RAM 58.0 · Disk 104.3 |
| GGUF4bit | 3.0 GB RAM 32.0 · Disk 39.3 | 7.7 GB RAM 32.0 · Disk 65.3 | 38.6 GB RAM 58.0 · Disk 104.3 |
GPU Recommendations
1× RTX 5080
16 GB VRAM
≈ 5647.1 tok/s
Effective production throughput ≈ 3953 tok/s
score 0.6233
1× RTX 5060 Ti
8 GB VRAM
≈ 2635.3 tok/s
Effective production throughput ≈ 1844.7 tok/s
score 0.639
8× H200 SXM
1128 GB VRAM
≈ 90352.9 tok/s
Effective production throughput ≈ 63247 tok/s
score 0.2197
1× RTX 5070
12 GB VRAM
≈ 3952.9 tok/s
Effective production throughput ≈ 2767 tok/s
score 0.6648
💬 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