Phind-CodeLlama 34B
Phind/Phind-CodeLlama-34B-v2
Parameters
34B
Context
16K
Downloads
500
Likes
834
Architecture
transformer
GPU Requirements
VRAM / RAM / disk estimates by quantization and usage scenario
| Quantization | Minimum | Recommended | Production |
|---|---|---|---|
| FP1616bit | 85.1 GB RAM 127.6 · Disk 137.8 | 104.5 GB RAM 156.7 · Disk 163.8 | 205.0 GB RAM 307.5 · Disk 202.8 |
| FP88bit | 45.0 GB RAM 67.5 · Disk 88.4 | 62.6 GB RAM 93.9 · Disk 114.4 | 161.3 GB RAM 242.0 · Disk 153.4 |
| INT88bit | 45.0 GB RAM 67.5 · Disk 88.4 | 62.6 GB RAM 93.9 · Disk 114.4 | 161.3 GB RAM 242.0 · Disk 153.4 |
| AWQ4bit | 25.7 GB RAM 38.6 · Disk 64.6 | 42.4 GB RAM 63.6 · Disk 90.6 | 140.3 GB RAM 210.4 · Disk 129.6 |
| GPTQ4bit | 26.0 GB RAM 39.0 · Disk 64.9 | 42.7 GB RAM 64.0 · Disk 90.9 | 140.5 GB RAM 210.8 · Disk 129.9 |
| GGUF4bit | 26.5 GB RAM 39.7 · Disk 65.6 | 43.2 GB RAM 64.8 · Disk 91.6 | 141.1 GB RAM 211.6 · Disk 130.6 |
GPU Recommendations
1× H100 SXM
80 GB VRAM
≈ 203.9 tok/s
Effective production throughput ≈ 142.7 tok/s
score 0.6261
2× RTX 3090
48 GB VRAM
≈ 91.2 tok/s
Effective production throughput ≈ 63.8 tok/s
score 0.6188
8× H200 SXM
1128 GB VRAM
≈ 934.9 tok/s
Effective production throughput ≈ 654.4 tok/s
score 0.2197
1× RTX A6000
48 GB VRAM
≈ 46.7 tok/s
Effective production throughput ≈ 32.7 tok/s
score 0.6543
💬 Community — real deployment experience
0 Articles · 0 Benchmarks
GPUs that fit (single card)
RTX A6000
48 GB VRAM
$265
/mo · vast
L40S
48 GB VRAM
$341
/mo · vast
A100 PCIE
80 GB VRAM
$390
/mo · vast
A100 SXM4
80 GB VRAM
$440
/mo · vast
RTX PRO 6000 Max-Q
48 GB VRAM
$465
/mo · vast