Update MTP doc (#4290)

docs/backend/speculative_decoding.ipynb

@@ -8,7 +8,7 @@
     "\n",
     "SGLang now provides an EAGLE2-based speculative decoding option. Our implementation aims to maximize speed and efficiency and is considered to be among the fastest in open-source LLM engines.\n",
     "\n",
-    "**Note:** Currently, Speculative Decoding in SGLang does not support radix cache.\n",
+    "**Note:** Currently, Speculative Decoding in SGLang is compatible with radix cache and chunked prefill.\n",
     "\n",
     "### Performance Highlights\n",
     "\n",

docs/references/deepseek.md

@@ -84,13 +84,13 @@ Please refer to [the example](https://github.com/sgl-project/sglang/tree/main/be
 
 - **Weight Absorption**: By applying the associative law of matrix multiplication to reorder computation steps, this method balances computation and memory access and improves efficiency in the decoding phase.
 
-- **Flashinfer MLA Wrapper**: By providing `--enable-flashinfer-mla` argument, the server will use MLA kernels customized by Flashinfer. More details can be referred to [this document](https://docs.flashinfer.ai/api/mla.html). Under long input scenarios, flashinfer mla can improve performance significantly. Optimized triton kernels will be used when flashinfer mla is turned off. Currently when using flashinfer mla wrapper and speculative decoding together, the `speculative_eagle_topk` parameter should be set to 1.
+- **Flashinfer MLA Wrapper**: By providing `--enable-flashinfer-mla` argument, the server will use MLA kernels customized by Flashinfer. More details can be referred to [this document](https://docs.flashinfer.ai/api/mla.html). Under long input scenarios, flashinfer mla can improve performance significantly. Optimized triton kernels will be used when flashinfer mla is turned off.
 
 - **FP8 Quantization**: W8A8 FP8 and KV Cache FP8 quantization enables efficient FP8 inference. Additionally, we have implemented Batched Matrix Multiplication (BMM) operator to facilitate FP8 inference in MLA with weight absorption.
 
 - **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.
 
-Overall, with these optimizations, we have achieved up to a 7x acceleration in output throughput compared to the previous version.
+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">
@@ -108,12 +108,14 @@ Overall, with these optimizations, we have achieved up to a 7x acceleration in o
   <img src="https://lmsys.org/images/blog/sglang_v0_4/dp_attention.svg" alt="Data Parallelism Attention for DeepSeek Series Models">
 </p>
 
-**Usage**: This optimization is aimed at improving throughput and should be used for scenarios with high QPS (Queries Per Second). Data Parallelism Attention optimization can be enabled by `--enable-dp-attention` for DeepSeek Series Models.
+With data parallelism attention enabled, we have achieved up to **1.9x** decoding throughput improvement compared to the previous version.
 
 <p align="center">
   <img src="https://lmsys.org/images/blog/sglang_v0_4/deepseek_coder_v2.svg" alt="Data Parallelism Attention Performance Comparison">
 </p>
 
+**Usage**: This optimization is aimed at improving throughput and should be used for scenarios with high QPS (Queries Per Second). Data Parallelism Attention optimization can be enabled by `--enable-dp-attention` for DeepSeek Series Models.
+
 **Reference**: Check [Blog](https://lmsys.org/blog/2024-12-04-sglang-v0-4/#data-parallelism-attention-for-deepseek-models).
 
 ### Multi Node Tensor Parallelism
@@ -130,7 +132,21 @@ Overall, with these optimizations, we have achieved up to a 7x acceleration in o
 
 - **Weight**: Per-128x128-block quantization for better numerical stability.
 
-**Usage**: turn on by default for DeepSeek V3 models.
+**Usage**: Turn on by default for DeepSeek V3 models.
+
+### Multi-token Prediction
+**Description**: SGLang implements DeepSeek V3 Multi-Token Prediction (MTP) based on [EAGLE speculative decoding](https://docs.sglang.ai/backend/speculative_decoding.html#EAGLE-Decoding). With this optimization, the decoding speed can be improved by **1.8x** for batch size 1 and **1.5x** for batch size 32 respectively on H200 TP8 setting.
+
+**Usage**:
+Add arguments `--speculative-algorithm`, `--speculative-draft-model-path`,
+`--speculative-num-steps`, `--speculative-eagle-topk` and `--speculative-num-draft-tokens` to enable this feature. For example:
+```
+python3 -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3 --speculative-algorithm EAGLE --speculative-draft-model-path lmsys/DeepSeek-V3-NextN --speculative-num-steps 1 --speculative-eagle-topk 1 --speculative-num-draft-tokens 2 --trust-remote --tp 8
+```
+- The draft model are available at huggingface: [lmsys/DeepSeek-V3-NextN](https://huggingface.co/lmsys/DeepSeek-V3-NextN), [lmsys/DeepSeek-R1-NextN](https://huggingface.co/lmsys/DeepSeek-R1-NextN). It can also be exported from original DeepSeek-V3/R1 model with [export_deepseek_nextn.py](https://github.com/sgl-project/sglang/blob/main/scripts/export_deepseek_nextn.py) script.
+- The best configuratin for `--speculative-num-steps`, `--speculative-eagle-topk` and `--speculative-num-draft-tokens` can be searched with [bench_speculative.py](https://github.com/sgl-project/sglang/blob/main/scripts/playground/bench_speculative.py) script for given batch size. The minimum configuration is `--speculative-num-steps 1 --speculative-eagle-topk 1 --speculative-num-draft-tokens 2`, which can achieve speedup for larger batch sizes.
+- Currently when using flashinfer mla wrapper (`--enable-flashinfer-mla`) and speculative decoding together, the `--speculative-eagle-topk` parameter should be set to `1`.
+
 
 ### Reasoning Content for DeepSeek R1