NVIDIA H200 141GBBudget pickB59.5 tok/s decode
~$30,000 MSRP
Can it run?
Runs well
Using Q4_K_M in vLLM
Fit status
Runs well
Decode
34.0 tok/s
TTFT
5690 ms
Safe context
20K
Memory
100.3 GB / 128.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Tight fit | 36.9 tok/s | 7639 ms | 35K |
| Chat | C | Runs well | 36.9 tok/s | 2865 ms | 11K |
| Coding | C | Runs well | 34.0 tok/s | 5690 ms | 20K |
| RAG | C | Tight fit | 36.9 tok/s | 9549 ms | 35K |
| Reasoning | C | Runs well | 36.9 tok/s | 6207 ms | 20K |
How Command A 111B (111B params) fits at each quantization level on AMD Instinct MI250X 128GB (128.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 43.3 GB | Low | D37 |
Q3_K_S | 3 | 54.4 GB | Low | D39 |
NVFP4 | 4 |
ollama run command-a-111bhuggingface-cli download command-a-111bUpgrade options
NVIDIA H200 141GBBudget pick~$30,000 MSRP
NVIDIA H200 PCIe 141GBBest value~$30,000 MSRP
NVIDIA B200 180GBBiggest leap~$30,000 MSRP
62.2 GB |
| Medium |
| D40 |
Q4_K_M | 4 | 67.7 GB | Medium | C41 |
Q5_K_M | 5 | 79.9 GB | High | C43 |
Q6_KBest for your GPU | 6 | 91.0 GB | High | C44 |
Q8_0 | 8 | 118.8 GB | Very High | C44 |
F16 | 16 | 227.6 GB | Maximum | F0 |