Support MHA with chunked prefix cache for DeepSeek chunked prefill (#5113)

docs/references/deepseek.md

@@ -92,13 +92,15 @@ Please refer to [the example](https://github.com/sgl-project/sglang/tree/main/be
 
 - **CUDA Graph & Torch.compile**: Both MLA and Mixture of Experts (MoE) are compatible with CUDA Graph and Torch.compile, which reduces latency and accelerates decoding speed for small batch sizes.
 
+- **Chunked Prefix Cache**: Chunked prefix cache optimization can increase throughput by cutting prefix cache into chunks, processing them with multi-head attention and merging their states. Its improvement can be significant when doing chunked prefill on long sequences. Currently this optimization is only available for FlashAttention3 backend.
+
 Overall, with these optimizations, we have achieved up to **7x** acceleration in output throughput compared to the previous version.
 
 <p align="center">
   <img src="https://lmsys.org/images/blog/sglang_v0_3/deepseek_mla.svg" alt="Multi-head Latent Attention for DeepSeek Series Models">
 </p>
 
-**Usage**: MLA optimization is enabled by default, to disable, use `--disable-mla`.
+**Usage**: MLA optimization is enabled by default. To disable MLA usage, use `--disable-mla`. To disable chunked prefix cache feature for mla, use `disable-chunked-prefix-cache`.
 
 **Reference**: Check [Blog](https://lmsys.org/blog/2024-09-04-sglang-v0-3/#deepseek-multi-head-latent-attention-mla-throughput-optimizations) and [Slides](https://github.com/sgl-project/sgl-learning-materials/blob/main/slides/lmsys_1st_meetup_deepseek_mla.pdf) for more details.