Can I run Kimi-Dev-72B on a GeForce RTX 2080 Ti?
Not at these settings. No indexed quantization of Kimi-Dev-72B fits GeForce RTX 2080 Ti at any context we compute, with q4_0 KV. The smallest shipped quantization is 21.47 GiB in weights alone, against 10.23 GiB usable. CPU offload can still run it, slowly.
Every quantization at every context
| Quant | Weights● | 4K◐ | 8K◐ | 16K◐ | 32K◐ | 64K◐ | 128K◐ |
|---|---|---|---|---|---|---|---|
| BF16 | 135.44 GiB | 136.7 | 137.1 | 137.8 | 139.2 | 142.0 | 147.6 |
| Q8_0 | 71.96 GiB | 73.2 | 73.6 | 74.3 | 75.7 | 78.5 | 84.1 |
| Q6_K | 59.93 GiB | 61.2 | 61.6 | 62.3 | 63.7 | 66.5 | 72.1 |
| Q4_1 | 42.56 GiB | 43.8 | 44.2 | 44.9 | 46.3 | 49.1 | 54.7 |
| Q4_0 | 38.54 GiB | 39.8 | 40.2 | 40.9 | 42.3 | 45.1 | 50.7 |
| IQ4_NL | 38.48 GiB | 39.8 | 40.1 | 40.8 | 42.2 | 45.0 | 50.7 |
| IQ4_XS | 37.02 GiB | 38.3 | 38.7 | 39.4 | 40.8 | 43.6 | 49.2 |
| Q3_K_M | 35.11 GiB | 36.4 | 36.7 | 37.4 | 38.9 | 41.7 | 47.3 |
| Q3_K_S | 32.12 GiB | 33.4 | 33.8 | 34.5 | 35.9 | 38.7 | 44.3 |
| UD-IQ3_XXS | 29.67 GiB | 31.0 | 31.3 | 32.0 | 33.4 | 36.2 | 41.8 |
| UD-IQ2_M | 27.56 GiB | 28.8 | 29.2 | 29.9 | 31.3 | 34.1 | 39.7 |
| UD-IQ2_XXS | 23.94 GiB | 25.2 | 25.6 | 26.3 | 27.7 | 30.5 | 36.1 |
| UD-IQ1_M | 22.35 GiB | 23.6 | 24.0 | 24.7 | 26.1 | 28.9 | 34.5 |
| UD-IQ1_S | 21.47 GiB | 22.7 | 23.1 | 23.8 | 25.2 | 28.0 | 33.6 |
Figures are GiB of total memory: weights plus KV cache plus compute buffer and backend overhead. Weights and KV are near-exact; the overhead term is modeled. Hover any cell for the breakdown.
Why other calculators disagree
A parameters × bits ÷ 8 estimate ignores two things that dominate at long context. First, the weights themselves are not the nominal rate — quantizations are mixtures, so the real file is consistently larger than the label implies. Second, most of this model's layers cache only a 131,072-token window rather than the full context.