Can it run?
Runs well
Using Q4_K_M in vLLM
Fit status
Runs well
Decode
90.8 tok/s
TTFT
2517 ms
Safe context
42K
Memory
110.9 GB / 288.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 92.1 tok/s | 3059 ms | 72K |
| Chat | C | Runs well | 92.1 tok/s | 1147 ms | 22K |
| Coding | C | Runs well | 90.8 tok/s | 2517 ms | 42K |
| RAG | C | Runs well | 92.1 tok/s | 3824 ms | 72K |
| Reasoning | C | Runs well | 92.1 tok/s | 2485 ms | 42K |
How Command R+ 104B (104B params) fits at each quantization level on AMD Instinct MI350X 288GB (288.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 40.6 GB | Low | D32 |
Q3_K_S | 3 | 51.0 GB | Low | D33 |
NVFP4 | 4 |
ollama run command-r-plus-104bhuggingface-cli download command-r-plus-104b58.2 GB |
| Medium |
| D34 |
Q4_K_M | 4 | 63.4 GB | Medium | D34 |
Q5_K_M | 5 | 74.9 GB | High | D35 |
Q6_K | 6 | 85.3 GB | High | D36 |
Q8_0 | 8 | 111.3 GB | Very High | D38 |
F16Best for your GPU | 16 | 213.2 GB | Maximum | C44 |