Skip to main content
Search

DeepSeek-AI

2 related articles

Subscribe to updates
  1. arXiv

    DeepSeek-V4.1-Flash cuts per-token global KV cache to 890 bytes, about a quarter of its predecessor

    DeepSeek-AI released DeepSeek-V4.1-Flash, a multimodal mixture-of-experts model aimed at the KV cache cost of long-context agent workloads. The backbone has 552B parameters and supports contexts up to one million tokens, activating 16B parameters per token during decode and only 8B during prefill. Through cross-layer KV reuse in CSA2 and FP4 caching, its global KV cache held in HBM falls to 890 bytes per token, roughly a quarter of DeepSeek-V4-Flash; with SWA Bounded Replay, the persistent KV cache is about one eighth of that baseline, while overall performance is better.

  2. arXiv

    DeepSeek-V4 preview supports million-token context: single-token inference FLOPs at 27% of V3.2 and KV cache at 10%

    DeepSeek-AI released a preview of the DeepSeek-V4 series, with DeepSeek-V4-Pro at 1.6T parameters (49B activated) and DeepSeek-V4-Flash at 284B parameters (13B activated), both supporting a one-million-token context length. Standard attention grows quadratically with sequence length, making ultra-long contexts a bottleneck for inference and test-time scaling; the series instead interleaves compressed sparse attention with heavily compressed attention and pairs it with low-precision storage. At the one-million-token setting, DeepSeek-V4-Pro needs 27% of the single-token inference FLOPs and 10% of the KV cache of DeepSeek-V3.2, while DeepSeek-V4-Flash needs 10% and 7%.

You’re all caught up