Qwen on NVIDIA Hopper
Can I run Qwen3 235B-A22B on an H100 SXM 80GB?
Not at Q4_K_M — it needs 135.1 GB against 74.4 GB available. You would need 2 of these cards.
Won't fit
135.1 GB of 74.4 GB · 182%0141 GB
Weights
132.8 GB
KV cache
1.5 GB
Runtime overhead
0.8 GB
Over the limit
60.7 GB
Short by 60.7 GB. You can run it with 51 of 94 layers on the H100 SXM 80GB and the rest in system RAM, at roughly 6.80 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.
Generation6.80tok/s
Prompt processing9381tok/s
Max context0tokens
KV per 1K tokens0GB
Every quantisation of Qwen3 235B-A22B on a H100 SXM 80GB
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 | — | 1.25 |
| INT8 / W8A8 | 8.50 | 232.3 GB | 234.6 GB | Won't fit | — | 2.77 |
| Q8_0 (GGUF) | 8.50 | 232.3 GB | 234.6 GB | Won't fit | — | 2.77 |
| FP8 (E4M3) | 8.00 | 218.9 GB | 221.2 GB | Won't fit | — | 2.97 |
| Q6_K | 6.56 | 179.5 GB | 181.8 GB | Won't fit | — | 3.99 |
| Q5_K_M | 5.67 | 155.2 GB | 157.5 GB | Won't fit | — | 5.06 |
| Q5_K_S | 5.52 | 151.1 GB | 153.4 GB | Won't fit | — | 5.28 |
| Q4_K_M | 4.85 | 132.8 GB | 135.1 GB | Won't fit | — | 6.80 |
| Q4_K_S | 4.58 | 125.5 GB | 127.8 GB | Won't fit | — | 7.63 |
| Q4_0 | 4.55 | 124.7 GB | 127.0 GB | Won't fit | — | 7.68 |
| AWQ 4-bit | 4.25 | 118.0 GB | 120.3 GB | Won't fit | — | 8.63 |
| GPTQ 4-bit | 4.25 | 118.0 GB | 120.3 GB | Won't fit | — | 8.63 |
| MXFP4 | 4.25 | 118.0 GB | 120.3 GB | Won't fit | — | 8.63 |
| IQ4_XS | 4.25 | 116.5 GB | 118.8 GB | Won't fit | — | 8.94 |
| Q3_K_M | 3.91 | 107.2 GB | 109.5 GB | Won't fit | — | 10.9 |
| IQ3_M | 3.70 | 101.5 GB | 103.8 GB | Won't fit | — | 12.5 |
| IQ3_XXS | 3.06 | 84.1 GB | 86.4 GB | Won't fit | — | 24.7 |
| Q2_K | 2.63 | 72.4 GB | 74.7 GB | Won't fit | 6K | 74.1 |
| IQ2_XXS | 2.06 | 56.9 GB | 59.2 GB | Runs comfortably | 91K | 90.8 |
| IQ1_M | 1.75 | 48.4 GB | 50.7 GB | Runs comfortably | 128K | 94.2 |