Can it run?
Runs well
Using Q4_K_M in vLLM
Fit status
Runs well
Decode
153.0 tok/s
TTFT
1265 ms
Safe context
33K
Memory
75.6 GB / 180.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 153.0 tok/s | 1840 ms | 33K |
| Chat | C | Runs well | 153.0 tok/s | 690 ms | 21K |
| Coding | C | Runs well | 153.0 tok/s | 1265 ms | 33K |
| RAG | C | Runs well | 153.0 tok/s | 2301 ms | 33K |
| Reasoning | C | Runs well | 153.0 tok/s | 1495 ms | 33K |
How Qwen 2.5 VL 72B (72B params) fits at each quantization level on NVIDIA B200 180GB (180.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 28.1 GB | Low | D33 |
Q3_K_S | 3 | 35.3 GB | Low | D34 |
NVFP4 | 4 | 40.3 GB | Medium | D34 |
Q4_K_M | 4 | 43.9 GB | Medium | D35 |
Q5_K_M | 5 | 51.8 GB | High | D36 |
Q6_K | 6 | 59.0 GB | High | D36 |
Q8_0 | 8 | 77.0 GB | Very High | D39 |
F16Best for your GPU | 16 | 147.6 GB | Maximum | C44 |
ollama run qwen-2.5-vl-72bhuggingface-cli download qwen-2.5-vl-72b