Can it run?
Runs well
Using Q4_K_M in vLLM
Fit status
Runs well
Decode
99.2 tok/s
TTFT
1951 ms
Safe context
27K
Memory
105.5 GB / 180.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | B | Runs well | 99.2 tok/s | 2837 ms | 47K |
| Chat | C | Runs well | 99.2 tok/s | 1064 ms | 15K |
| Coding | C | Runs well | 99.2 tok/s | 1951 ms | 27K |
| RAG | B | Runs well | 99.2 tok/s | 3547 ms | 47K |
| Reasoning | C | Runs well | 99.2 tok/s | 2305 ms | 27K |
How Command A 111B (111B params) fits at each quantization level on NVIDIA B200 180GB (180.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 43.3 GB | Low | D35 |
Q3_K_S | 3 | 54.4 GB | Low | D36 |
NVFP4 | 4 | 62.2 GB | Medium | D37 |
Q4_K_M | 4 | 67.7 GB | Medium | D37 |
Q5_K_M | 5 | 79.9 GB | High | D39 |
Q6_K | 6 | 91.0 GB | High | C40 |
Q8_0Best for your GPU | 8 | 118.8 GB | Very High | C44 |
F16 | 16 | 227.6 GB | Maximum | F0 |
ollama run command-a-111bhuggingface-cli download command-a-111b