Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
51.4 tok/s
TTFT
3766 ms
Safe context
22K
Memory
22.8 GB / 32.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Tight fit | 51.4 tok/s | 5478 ms | 39K |
| Chat | C | Runs well | 51.4 tok/s | 2054 ms | 12K |
| Coding | B | Runs well | 51.4 tok/s | 3766 ms | 22K |
| RAG | C | Tight fit | 51.4 tok/s | 6847 ms | 39K |
| Reasoning | B | Runs well | 51.4 tok/s | 4451 ms | 22K |
How Mistral Small 3.2 24B (24B params) fits at each quantization level on RTX PRO 4500 Blackwell 32GB (32.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 9.4 GB | Low | D35 |
Q3_K_S | 3 | 11.8 GB | Low | D37 |
NVFP4 | 4 | 13.4 GB | Medium | D38 |
Q4_K_M | 4 | 14.6 GB | Medium | D39 |
Q5_K_M | 5 | 17.3 GB | High | C41 |
Q6_K | 6 | 19.7 GB | High | C42 |
Q8_0Best for your GPU | 8 | 25.7 GB | Very High | C44 |
F16 | 16 | 49.2 GB | Maximum | F0 |
ollama run mistral-small-3.2-24bhuggingface-cli download mistral-small-3.2-24b