Can it run?
Very compromised
Using Q4_K_M in vLLM
Fit status
Very compromised
Decode
49.0 tok/s
TTFT
3949 ms
Safe context
14K
Memory
329.5 GB / 288.0 GB
Offload
10%
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Very compromised | 48.4 tok/s | 5823 ms | 28K |
| Chat | C | Very compromised | 49.4 tok/s | 2139 ms | 7K |
| Coding | C | Very compromised | 49.0 tok/s | 3949 ms | 14K |
| RAG | C | Very compromised | 48.4 tok/s | 7279 ms | 28K |
| Reasoning | C | Very compromised | 49.0 tok/s | 4667 ms | 14K |
How Qwen3-Coder 480B A35B Instruct (480B params) fits at each quantization level on AMD Instinct MI350X 288GB (288.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_KBest for your GPU | 2 | 187.2 GB | Low | C43 |
Q3_K_S | 3 | 235.2 GB | Low | C45 |
huggingface-cli download qwen-3-coder-480b-a35bNVFP4 |
| 4 |
268.8 GB |
| Medium |
| C45 |
Q4_K_M | 4 | 292.8 GB | Medium | F0 |
Q5_K_M | 5 | 345.6 GB | High | F0 |
Q6_K | 6 | 393.6 GB | High | F0 |
Q8_0 | 8 | 513.6 GB | Very High | F0 |
F16 | 16 | 984.0 GB | Maximum | F0 |