Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
57.4 tok/s
TTFT
3375 ms
Safe context
22K
Memory
29.7 GB / 40.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Tight fit | 66.9 tok/s | 4208 ms | 37K |
| Chat | B | Runs well | 66.9 tok/s | 1578 ms | 12K |
| Coding | B | Runs well | 57.4 tok/s | 3375 ms | 22K |
| RAG | C | Tight fit | 66.9 tok/s | 5260 ms | 37K |
| Reasoning | B | Runs well | 66.9 tok/s | 3419 ms | 22K |
How Qwen 2.5 Coder 32B (32B params) fits at each quantization level on NVIDIA A100 40GB (40.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 12.5 GB | Low | D36 |
Q3_K_S | 3 | 15.7 GB | Low | D38 |
NVFP4 | 4 |
ollama run qwen-2.5-coder-32bhuggingface-cli download qwen-2.5-coder-32b17.9 GB |
| Medium |
| D39 |
Q4_K_M | 4 | 19.5 GB | Medium | D40 |
Q5_K_M | 5 | 23.0 GB | High | C42 |
Q6_KBest for your GPU | 6 | 26.2 GB | High | C43 |
Q8_0 | 8 | 34.2 GB | Very High | C44 |
F16 | 16 | 65.6 GB | Maximum | F0 |