Can it run?
Runs well
Using Q4_K_M in vLLM
Fit status
Runs well
Decode
63.6 tok/s
TTFT
3046 ms
Safe context
23K
Memory
96.2 GB / 141.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | B | Runs well | 63.6 tok/s | 4431 ms | 40K |
| Chat | C | Runs well | 63.6 tok/s | 1662 ms | 13K |
| Coding | B | Runs well | 63.6 tok/s | 3046 ms | 23K |
| RAG | B | Runs well | 63.6 tok/s | 5538 ms | 40K |
| Reasoning | B | Runs well | 63.6 tok/s | 3600 ms | 23K |
How Command R+ 104B (104B params) fits at each quantization level on NVIDIA H200 PCIe 141GB (141.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 40.6 GB | Low | D35 |
Q3_K_S | 3 | 51.0 GB | Low | D37 |
NVFP4 | 4 | 58.2 GB | Medium | D38 |
Q4_K_M | 4 | 63.4 GB | Medium | D39 |
Q5_K_M | 5 | 74.9 GB | High | C41 |
Q6_K | 6 | 85.3 GB | High | C42 |
Q8_0Best for your GPU | 8 | 111.3 GB | Very High | C44 |
F16 | 16 | 213.2 GB | Maximum | F0 |
ollama run command-r-plus-104bhuggingface-cli download command-r-plus-104b