How much memory does DeepSeek V4 need in FP8?
DeepSeek V4 has 1600 billion parameters. The FP8 format decides how much memory is needed to load them.
How is this footprint calculated?
DeepSeek V4 totals 1600 billion parameters. The FP8 format takes 1.0 byte per parameter. The product gives the weights, to which a 15% runtime margin is added for activations and buffers. The attention cache is not included: it grows with context and concurrency.
What does the format change?
FP8 brings negligible loss on most tasks.
| Format | Weights in memory | Compatible platforms |
|---|---|---|
| FP16 | 3,680 GB | 1 |
| FP8 | 1,840 GB | 2 |
| NVFP4 | 920 GB | 4 |
| Q4 | 920 GB | 4 |
Which platforms qualify?
See the DeepSeek V4 fact sheet.
How much memory for DeepSeek V4 in FP8?
About 1,840 GB for the weights, excluding the attention cache.
Which format should I choose for DeepSeek V4?
FP8 brings negligible loss on most tasks. The most precise format that fits the target machine remains the best choice.
Method
Parameter counts and memory figures come from the site fact sheets. Memory footprints are calculated, not measured: a reading on real hardware may differ depending on the engine and the exact weight format. Prices are indicative and not contractual.