farbodtavakkoli/OTel-2.0-LLM-31B-IT
farbodtavakkoli/OTel-2.0-LLM-31B-IT
Parameters
22.2B
Context
256K
Downloads
5M
Likes
12
Architecture
gemma4_text
60 layers · 16 KV heads · head 256 · hidden 5376
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 64.1 GB RAM 96.1 · Disk 103.6 | 148.7 GB RAM 223.0 · Disk 129.6 | 730.0 GB RAM 1095.0 · Disk 168.6 |
| FP88bit | 37.9 GB RAM 56.9 · Disk 71.3 | 121.3 GB RAM 182.0 · Disk 97.3 | 701.5 GB RAM 1052.2 · Disk 136.3 |
| INT88bit | 37.9 GB RAM 56.9 · Disk 71.3 | 121.3 GB RAM 182.0 · Disk 97.3 | 701.5 GB RAM 1052.2 · Disk 136.3 |
| AWQ4bit | 25.3 GB RAM 38.0 · Disk 55.8 | 108.1 GB RAM 162.2 · Disk 81.8 | 687.7 GB RAM 1031.6 · Disk 120.8 |
| GPTQ4bit | 25.5 GB RAM 38.2 · Disk 56.0 | 108.3 GB RAM 162.5 · Disk 82.0 | 687.9 GB RAM 1031.8 · Disk 121.0 |
| GGUF4bit | 25.8 GB RAM 38.7 · Disk 56.4 | 108.7 GB RAM 163.0 · Disk 82.4 | 688.3 GB RAM 1032.4 · Disk 121.4 |
GPU Recommendations
4× H200 SXM
564 GB VRAM
≈ 1073.6 tok/s
Effective production throughput ≈ 751.5 tok/s
score 0.3205
8× RTX 4070S Ti
128 GB VRAM
≈ 0 tok/s (N/A — benchmark unavailable)
Effective production throughput ≈ 0 tok/s
score 0.5265
8× H200 SXM
1128 GB VRAM
≈ 1431.5 tok/s
Effective production throughput ≈ 1002.1 tok/s
score 0.2197
1× H200 SXM
141 GB VRAM
≈ 447.3 tok/s
Effective production throughput ≈ 313.1 tok/s
score 0.6665
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
H200 SXM
141 GB VRAM
$2,555
/mo · lambda
B200
180 GB VRAM
$3,013
/mo · vast
B200 SXM
180 GB VRAM
—