Can I run DeepSeek-V4-Flash-0731 on a 12GB card?

No. No published quantization of DeepSeek-V4-Flash-0731 fits a 12GB card at any context with f16 KV cache.

How much VRAM does DeepSeek-V4-Flash-0731 need?

Its published quantizations range from 76.87 GiB to 145.64 GiB in weights alone, before KV cache and runtime overhead.

Does DeepSeek-V4-Flash-0731 fit in 12GB of VRAM?

Not at these settings. No indexed quantization of DeepSeek-V4-Flash-0731 fits 12GB card at any context we compute, with f16 KV. The smallest shipped quantization is 76.87 GiB in weights alone, against 11.16 GiB usable. CPU offload can still run it, slowly.

From the file· weights summed from filesFrom the file· KV computed per layerPredicted· speed and compute buffer

KV cache dtypef16 q8_0 q4_0— 12 GB card, 11.16 GiB usable after overhead

Every quantization at every context

total memory required; green fits, red does not

Quant	Weights●	4K◐	8K◐	16K◐	32K◐	64K◐	128K◐
MXFP4	145.64 GiB	146.5	146.5	146.5	146.5	146.5	146.5
UD-IQ4_NL	127.28 GiB	128.2	128.2	128.2	128.2	128.2	128.2
UD-IQ4_XS	127.28 GiB	128.2	128.2	128.2	128.2	128.2	128.2
UD-Q3_K_M	119.28 GiB	120.2	120.2	120.2	120.2	120.2	120.2
UD-IQ3_S	108.10 GiB	109.0	109.0	109.0	109.0	109.0	109.0
UD-IQ3_XXS	97.05 GiB	98.0	98.0	98.0	98.0	98.0	98.0
UD-IQ2_M	84.68 GiB	85.6	85.6	85.6	85.6	85.6	85.6
UD-IQ2_XXS	84.62 GiB	85.5	85.5	85.5	85.5	85.5	85.5
UD-IQ1_M	80.93 GiB	81.8	81.8	81.8	81.8	81.8	81.8
UD-IQ1_S	76.87 GiB	77.8	77.8	77.8	77.8	77.8	77.8

●From the file◐Predictedwhat these mean

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 there are no speeds on this page

A capacity is not a card. Whether a model fits depends only on memory, so every figure above holds for any 12GB accelerator. How fast it runs depends on memory bandwidth, which varies several-fold between cards of the same capacity — so putting a tokens-per-second number here would be inventing one. Pick a specific card from hardware and the speed column appears.

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 128-token window rather than the full context.