EXAONE 3.5 7.8B
LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct
Parameters
7.8B
Context
128K
Downloads
519.1K
Likes
159
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 23.3 GB RAM 35.0 · Disk 61.7 | 39.9 GB RAM 59.9 · Disk 87.7 | 137.7 GB RAM 206.5 · Disk 126.7 |
| FP88bit | 14.1 GB RAM 32.0 · Disk 50.3 | 30.3 GB RAM 45.5 · Disk 76.3 | 127.6 GB RAM 191.4 · Disk 115.3 |
| INT88bit | 14.1 GB RAM 32.0 · Disk 50.3 | 30.3 GB RAM 45.5 · Disk 76.3 | 127.6 GB RAM 191.4 · Disk 115.3 |
| AWQ4bit | 9.7 GB RAM 32.0 · Disk 44.9 | 25.7 GB RAM 38.5 · Disk 70.9 | 122.8 GB RAM 184.2 · Disk 109.9 |
| GPTQ4bit | 9.8 GB RAM 32.0 · Disk 45.0 | 25.7 GB RAM 38.6 · Disk 71.0 | 122.9 GB RAM 184.3 · Disk 110.0 |
| GGUF4bit | 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 |
GPU Recommendations
1× RTX 5090
32 GB VRAM
≈ 475.3 tok/s
Effective production throughput ≈ 332.7 tok/s
score 0.6837
2× Tesla V100
32 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.6331
8× H200 SXM
1128 GB VRAM
≈ 4074.3 tok/s
Effective production throughput ≈ 2852 tok/s
score 0.2197
1× RTX A6000
48 GB VRAM
≈ 203.7 tok/s
Effective production throughput ≈ 142.6 tok/s
score 0.6338
💬 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