Instinct MI300X 192GB vs M5 Max 64GB
For running local LLMs · 1 report across 1 model
Tokens per second by model
| Model | Instinct MI300X 192GB | M5 Max 64GB |
|---|---|---|
| Qwen3.6up to 35B | — | 32.0n=1 |
For running local LLMs · 1 report across 1 model
| Model | Instinct MI300X 192GB | M5 Max 64GB |
|---|---|---|
| Qwen3.6up to 35B | — | 32.0n=1 |