Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
214.6 tok/s
TTFT
902 ms
Safe context
35K
Memory
64.9 GB / 141.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 250.4 tok/s | 1125 ms | 69K |
| Chat | C | Runs well | 250.4 tok/s | 422 ms | 17K |
| Coding | C | Runs well | 214.6 tok/s | 902 ms | 35K |
| RAG | C | Runs well | 250.4 tok/s | 1406 ms | 69K |
| Reasoning | C | Runs well | 250.4 tok/s | 914 ms | 35K |
How Qwen3-Coder-Next (80B params) fits at each quantization level on NVIDIA H200 PCIe 141GB (141.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 31.2 GB | Low | D34 |
Q3_K_S | 3 | 39.2 GB | Low | D36 |
NVFP4 | 4 |
ollama run qwen-3-coder-nexthuggingface-cli download qwen-3-coder-next44.8 GB |
| Medium |
| D36 |
Q4_K_M | 4 | 48.8 GB | Medium | D37 |
Q5_K_M | 5 | 57.6 GB | High | D38 |
Q6_K | 6 | 65.6 GB | High | D40 |
Q8_0Best for your GPU | 8 | 85.6 GB | Very High | C43 |
F16 | 16 | 164.0 GB | Maximum | F0 |