Can I run this LLM?
Can a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB) run Qwen3 Coder 30B-A3B (MoE)?
Qwen · Apache 2.0 · estimated for a 288.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Qwen3 Coder 30B-A3B (MoE) at FP16 · 61.1 GB · 263–394 tok/s on a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 11.3 GB | 336–503 tok/s | Runs great |
| Q3_K_M | 14.7 GB | 329–494 tok/s | Runs great |
| Q4_K_M | 18.6 GB | 322–484 tok/s | Runs great |
| Q5_K_M | 21.7 GB | 317–476 tok/s | Runs great |
| Q6_K | 25.1 GB | 312–467 tok/s | Runs great |
| Q8_0 | 32.5 GB | 300–450 tok/s | Runs great |
| FP16 | 61.1 GB | 263–394 tok/s | Runs great |
Open the full calculator →Everything a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB) runsQwen3 Coder 30B-A3B (MoE) on other machinesWhat to buy to run it →
Ran Qwen3 Coder 30B-A3B (MoE) on a NVIDIA GB300 Grace Blackwell (288GB HBM + 480GB)? Add your real numbers →