meta-llama/Llama-3.1-8B-Instruct
meta-llama/Llama-3.1-8B-Instruct
Meta Llama 3.1 8B, the default open model baseline.
Parameters
8B
Context
128K
Downloads
7.6M
Likes
6.6K
Architecture
—
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 23.8 GB RAM 35.7 · Disk 62.3 | 40.4 GB RAM 60.6 · Disk 88.3 | 138.2 GB RAM 207.3 · Disk 127.3 |
| FP88bit | 14.4 GB RAM 32.0 · Disk 50.6 | 30.6 GB RAM 45.8 · Disk 76.6 | 127.9 GB RAM 191.8 · Disk 115.6 |
| INT88bit | 14.4 GB RAM 32.0 · Disk 50.6 | 30.6 GB RAM 45.8 · Disk 76.6 | 127.9 GB RAM 191.8 · Disk 115.6 |
| AWQ4bit | 9.8 GB RAM 32.0 · Disk 45.0 | 25.8 GB RAM 38.7 · Disk 71.0 | 122.9 GB RAM 184.4 · Disk 110.0 |
| GPTQ4bit | 9.9 GB RAM 32.0 · Disk 45.1 | 25.9 GB RAM 38.8 · Disk 71.1 | 123.0 GB RAM 184.5 · Disk 110.1 |
| GGUF4bit | 10.0 GB RAM 32.0 · Disk 45.3 | 26.0 GB RAM 39.0 · Disk 71.3 | 123.1 GB RAM 184.7 · Disk 110.3 |
GPU Recommendations
1× RTX 5090
32 GB VRAM
≈ 464.2 tok/s
Effective production throughput ≈ 324.9 tok/s
score 0.6826
2× Tesla V100
32 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.632
8× H200 SXM
1128 GB VRAM
≈ 3979.3 tok/s
Effective production throughput ≈ 2785.5 tok/s
score 0.2197
1× RTX A6000
48 GB VRAM
≈ 199 tok/s
Effective production throughput ≈ 139.3 tok/s
score 0.6345
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX A6000
48 GB VRAM
$210
/mo · vast
RTX 5090
32 GB VRAM
$244
/mo · vast
L40S
48 GB VRAM
$341
/mo · vast
RTX PRO 6000 WS
48 GB VRAM
$586
/mo · vast
A100 80GB
80 GB VRAM
$876
/mo · vast