Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
87.4 tok/s
TTFT
2214 ms
Safe context
43K
Memory
29.9 GB / 80.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 102.0 tok/s | 2761 ms | 75K |
| Chat | C | Runs well | 102.0 tok/s | 1035 ms | 23K |
| Coding | C | Runs well | 87.4 tok/s | 2214 ms | 43K |
| RAG | C | Runs well | 102.0 tok/s | 3451 ms | 75K |
| Reasoning | C | Runs well | 102.0 tok/s | 2243 ms | 43K |
How gemma 3 27b it (27B params) fits at each quantization level on NVIDIA H100 PCIe 80GB (80.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 10.5 GB | Low | D32 |
Q3_K_S | 3 | 13.2 GB | Low | D33 |
NVFP4 | 4 |
15.1 GB |
| Medium |
| D33 |
Q4_K_M | 4 | 16.5 GB | Medium | D34 |
Q5_K_M | 5 | 19.4 GB | High | D35 |
Q6_K | 6 | 22.1 GB | High | D35 |
Q8_0 | 8 | 28.9 GB | Very High | D37 |
F16Best for your GPU | 16 | 55.4 GB | Maximum | C44 |