Can I run this LLM?
Can a AMD Instinct MI300X (192GB) run Qwen3 Coder Next 80B-A3B (MoE)?
Qwen · Apache 2.0 · estimated for a 192.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Qwen3 Coder Next 80B-A3B (MoE) at Q8_0 · 84.8 GB · 280–421 tok/s on a AMD Instinct MI300X (192GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 29.2 GB | 330–495 tok/s | Runs great |
| Q3_K_M | 38.3 GB | 321–481 tok/s | Runs great |
| Q4_K_M | 49.3 GB | 310–465 tok/s | Runs great |
| Q5_K_M | 59.2 GB | 301–452 tok/s | Runs great |
| Q6_K | 65.8 GB | 296–443 tok/s | Runs great |
| Q8_0 | 84.8 GB | 280–421 tok/s | Runs great |
| FP16 | 159.5 GB | 233–350 tok/s | Tight / slow |
Open the full calculator →Everything a AMD Instinct MI300X (192GB) runsQwen3 Coder Next 80B-A3B (MoE) on other machinesWhat to buy to run it →
Ran Qwen3 Coder Next 80B-A3B (MoE) on a AMD Instinct MI300X (192GB)? Add your real numbers →