Qwen on AMD RDNA3
Can I run Qwen3 235B-A22B on an Radeon RX 7900 XTX?
Not at Q4_K_M — it needs 135.1 GB against 21.8 GB available. You would need 7 of these cards.
Won't fit
135.1 GB of 21.8 GB · 621%0141 GB
Weights
132.8 GB
KV cache
1.5 GB
Runtime overhead
0.8 GB
Over the limit
113.4 GB
Short by 113.4 GB. You can run it with 13 of 94 layers on the Radeon RX 7900 XTX and the rest in system RAM, at roughly 3.07 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.
Generation3.07tok/s
Prompt processing1167tok/s
Max context0tokens
KV per 1K tokens0GB
Every quantisation of Qwen3 235B-A22B on a Radeon RX 7900 XTX
Highlighted row is the highest quality that still fits at 8K context.
| Quantisation | bpw | Weights | Total | Verdict | Max ctx | tok/s |
|---|---|---|---|---|---|---|
| FP16 / BF16 | 16.00 | 437.9 GB | 440.2 GB | Won't fit | — | 0.89 |
| INT8 / W8A8 | 8.50 | 232.3 GB | 234.6 GB | Won't fit | — | 1.70 |
| Q8_0 (GGUF) | 8.50 | 232.3 GB | 234.6 GB | Won't fit | — | 1.70 |
| FP8 (E4M3) | 8.00 | 218.9 GB | 221.2 GB | Won't fit | — | 1.82 |
| Q6_K | 6.56 | 179.5 GB | 181.8 GB | Won't fit | — | 2.24 |
| Q5_K_M | 5.67 | 155.2 GB | 157.5 GB | Won't fit | — | 2.60 |
| Q5_K_S | 5.52 | 151.1 GB | 153.4 GB | Won't fit | — | 2.69 |
| Q4_K_M | 4.85 | 132.8 GB | 135.1 GB | Won't fit | — | 3.07 |
| Q4_K_S | 4.58 | 125.5 GB | 127.8 GB | Won't fit | — | 3.27 |
| Q4_0 | 4.55 | 124.7 GB | 127.0 GB | Won't fit | — | 3.29 |
| AWQ 4-bit | 4.25 | 118.0 GB | 120.3 GB | Won't fit | — | 3.49 |
| GPTQ 4-bit | 4.25 | 118.0 GB | 120.3 GB | Won't fit | — | 3.49 |
| MXFP4 | 4.25 | 118.0 GB | 120.3 GB | Won't fit | — | 3.49 |
| IQ4_XS | 4.25 | 116.5 GB | 118.8 GB | Won't fit | — | 3.54 |
| Q3_K_M | 3.91 | 107.2 GB | 109.5 GB | Won't fit | — | 3.90 |
| IQ3_M | 3.70 | 101.5 GB | 103.8 GB | Won't fit | — | 4.14 |
| IQ3_XXS | 3.06 | 84.1 GB | 86.4 GB | Won't fit | — | 5.07 |
| Q2_K | 2.63 | 72.4 GB | 74.7 GB | Won't fit | — | 6.06 |
| IQ2_XXS | 2.06 | 56.9 GB | 59.2 GB | Won't fit | — | 8.13 |
| IQ1_M | 1.75 | 48.4 GB | 50.7 GB | Won't fit | — | 9.92 |