Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
110.7 tok/s
TTFT
1750 ms
Safe context
38K
Memory
33.7 GB / 80.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 124.5 tok/s | 2262 ms | 66K |
| Chat | C | Runs well | 124.5 tok/s | 848 ms | 20K |
| Coding | C | Runs well | 110.7 tok/s | 1750 ms | 38K |
| RAG | C | Runs well | 124.5 tok/s | 2828 ms | 66K |
| Reasoning | C | Runs well | 124.5 tok/s | 1838 ms | 38K |
How Qwen 2.5 Coder 32B (32B params) fits at each quantization level on NVIDIA H800 80GB (80.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 12.5 GB | Low | D33 |
Q3_K_S | 3 | 15.7 GB | Low | D34 |
NVFP4 | 4 |
ollama run qwen-2.5-coder-32bhuggingface-cli download qwen-2.5-coder-32b17.9 GB |
| Medium |
| D34 |
Q4_K_M | 4 | 19.5 GB | Medium | D35 |
Q5_K_M | 5 | 23.0 GB | High | D36 |
Q6_K | 6 | 26.2 GB | High | D36 |
Q8_0 | 8 | 34.2 GB | Very High | D39 |
F16Best for your GPU | 16 | 65.6 GB | Maximum | C44 |