Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
357.7 tok/s
TTFT
541 ms
Safe context
58K
Memory
79.6 GB / 288.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 357.7 tok/s | 787 ms | 116K |
| Chat | C | Runs well | 357.7 tok/s | 350 ms | 29K |
| Coding | C | Runs well | 357.7 tok/s | 541 ms | 58K |
| RAG | C | Runs well | 357.7 tok/s | 984 ms | 116K |
| Reasoning | C | Runs well | 357.7 tok/s | 640 ms | 58K |
How Qwen3-Coder-Next (80B params) fits at each quantization level on AMD Instinct MI350X 288GB (288.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 31.2 GB | Low | D32 |
Q3_K_S | 3 | 39.2 GB | Low | D32 |
NVFP4 | 4 |
ollama run qwen-3-coder-nexthuggingface-cli download qwen-3-coder-next44.8 GB |
| Medium |
| D33 |
Q4_K_M | 4 | 48.8 GB | Medium | D33 |
Q5_K_M | 5 | 57.6 GB | High | D34 |
Q6_K | 6 | 65.6 GB | High | D34 |
Q8_0 | 8 | 85.6 GB | Very High | D36 |
F16Best for your GPU | 16 | 164.0 GB | Maximum | C42 |