Can it run?
Runs well
Using Q4_K_M in vLLM
Fit status
Runs well
Decode
51.0 tok/s
TTFT
3793 ms
Safe context
22K
Memory
101.6 GB / 141.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Tight fit | 51.0 tok/s | 5517 ms | 38K |
| Chat | C | Runs well | 51.0 tok/s | 2069 ms | 12K |
| Coding | B | Runs well | 51.0 tok/s | 3793 ms | 22K |
| RAG | C | Tight fit | 51.0 tok/s | 6896 ms | 38K |
| Reasoning | B | Runs well | 51.0 tok/s | 4483 ms | 22K |
How Command A 111B (111B params) fits at each quantization level on NVIDIA H200 141GB (141.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 43.3 GB | Low | D36 |
Q3_K_S | 3 | 54.4 GB | Low | D38 |
NVFP4 | 4 |
ollama run command-a-111bhuggingface-cli download command-a-111b62.2 GB |
| Medium |
| D39 |
Q4_K_M | 4 | 67.7 GB | Medium | D40 |
Q5_K_M | 5 | 79.9 GB | High | C42 |
Q6_KBest for your GPU | 6 | 91.0 GB | High | C43 |
Q8_0 | 8 | 118.8 GB | Very High | C44 |
F16 | 16 | 227.6 GB | Maximum | F0 |