AMD Instinct MI350X 288GBBudget pick
C40.6 tok/s decode
~$8,000 MSRP
Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
30.0 tok/s
TTFT
6452 ms
Safe context
20K
Memory
207.6 GB / 256.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs with offload | 30.4 tok/s | 9256 ms | 34K |
| Chat | C | Runs well | 30.4 tok/s | 3471 ms | 11K |
| Coding | C | Runs well | 30.0 tok/s | 6452 ms | 20K |
| RAG | C | Runs with offload | 30.4 tok/s | 11569 ms | 34K |
| Reasoning | C | Runs well | 30.4 tok/s | 7520 ms | 20K |
How K EXAONE 236B A23B (236B params) fits at each quantization level on AMD Instinct MI325X 256GB (256.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 92.0 GB | Low | D37 |
Q3_K_S | 3 | 115.6 GB | Low | D39 |
NVFP4 | 4 |
Upgrade options
~$8,000 MSRP
132.2 GB |
| Medium |
| D40 |
Q4_K_M | 4 | 144.0 GB | Medium | C41 |
Q5_K_M | 5 | 169.9 GB | High | C43 |
Q6_KBest for your GPU | 6 | 193.5 GB | High | C44 |
Q8_0 | 8 | 252.5 GB | Very High | C44 |
F16 | 16 | 483.8 GB | Maximum | F0 |