Can Mixtral 8x22B run on AMD Instinct MI325X 256GB?
YES — Runs Great
Mixtral 8x22B needs ~115.9 GB VRAM. AMD Instinct MI325X 256GB has 256.0 GB. With Q4_K_M quantization, expect ~106 tok/s.
Operating mode
Choose the run profile you care about
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
Fit status
Runs well
Decode
105.8 tok/s
TTFT
1829 ms
Safe context
66K
Memory
115.9 GB / 256.0 GB
Memory breakdown
See how fast it feels
What limits this setup
This setup is broadly balanced for this model.
No major red flags
This recommendation has enough memory headroom and acceptable estimated speed for the selected workload.
Best improvement path
Performance by workload
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | B | Runs well | 105.8 tok/s | 998 ms | 66K |
| Coding | B | Runs well | 105.8 tok/s | 1829 ms | 66K |
| Agentic Coding | B | Runs well | 105.8 tok/s | 2661 ms | 66K |
| Reasoning | B | Runs well | 105.8 tok/s | 2162 ms | 66K |
| RAG | B | Runs well | 105.8 tok/s | 3326 ms | 66K |
Quantization options
How Mixtral 8x22B (141B params) fits at each quantization level on AMD Instinct MI325X 256GB (256.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 55.0 GB | Low | C54 |
Q3_K_S | 3 | 69.1 GB | Low | B55 |
NVFP4 | 4 |
Get started
Copy-paste commands to run Mixtral 8x22B on your machine.
Run
ollama run mixtral:8x22b