[Optimization] Auto set num_max_dispatch_tokens_per_rank

[RichardWooSJTU]

Motivation

当前低时延EP通信所需要的num_max_dispatch_tokens_per_rank参数只能通过model目录中的config.json指定，在max_num_seqs发生改变时还需要手动设置不太友好

💡 If this PR is a Cherry Pick, the PR title needs to follow the format by adding the [Cherry-Pick] label at the very beginning and appending the original PR ID at the end. For example, [Cherry-Pick][CI] Add check trigger and logic(#5191)

💡 如若此PR是Cherry Pick，PR标题需遵循格式，在最开始加上[Cherry-Pick]标签，以及最后面加上原PR ID，例如[Cherry-Pick][CI] Add check trigger and logic(#5191)

Modifications

考虑到当前需要改变这个参数的场景仅有

1. 关闭投机解码时，设置为max_num_seqs即可
2. 打开投机解码是，设置为max_num_seqs * (num_spec_tokens+1)

Usage or Command

Accuracy Tests

Checklist

• Add at least a tag in the PR title.
  • Tag list: [[FDConfig],[APIServer],[Engine], [Scheduler], [PD Disaggregation], [Executor], [Graph Optimization], [Speculative Decoding], [RL], [Models], [Quantization], [Loader], [OP], [KVCache], [DataProcessor], [BugFix], [Docs], [CI], [Optimization], [Feature], [Benchmark], [Others], [XPU], [HPU], [GCU], [DCU], [Iluvatar], [Metax]]
  • You can add new tags based on the PR content, but the semantics must be clear.
• Format your code, run pre-commit before commit.
• Add unit tests. Please write the reason in this PR if no unit tests.
• Provide accuracy results.
• If the current PR is submitting to the release branch, make sure the PR has been submitted to the develop branch, then cherry-pick it to the release branch with the [Cherry-Pick] PR tag.

[paddle-bot]

Thanks for your contribution!

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (develop@e0a1653). Learn more about missing BASE report.

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #7237   +/-   ##
==========================================
  Coverage           ?   73.52%           
==========================================
  Files              ?      383           
  Lines              ?    53644           
  Branches           ?     8421           
==========================================
  Hits               ?    39440           
  Misses             ?    11524           
  Partials           ?     2680           

Flag	Coverage Δ
GPU	73.52% <100.00%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[PaddlePaddle-bot]

🤖 AI Code Review | 2026-04-15

📋 Review 摘要

PR 概述：自动计算并设置 num_max_dispatch_tokens_per_rank 参数，根据 max_num_seqs 和投机解码配置动态调整

变更范围：fastdeploy/config.py

影响面 Tag：[FDConfig]

问题

级别	文件	概述
🔴 Bug	fastdeploy/config.py:2173	变量 num_spec_tokens 作用域错误，投机解码关闭时会导致 NameError

总体评价

PR 的优化意图合理，能够提升用户体验（无需手动同步配置），但存在一个阻塞性 Bug 需要修复：变量 num_spec_tokens 在投机解码关闭时未定义，会导致运行时错误。修复后建议添加单元测试覆盖各种场景。

[PaddlePaddle-bot]

🔴 Bug 变量作用域错误

num_spec_tokens 变量只在投机解码开启时（第 2161 行）才被定义，但在第 2173 行的日志中使用了该变量。当投机解码关闭时，num_spec_tokens 未定义会导致 NameError。

虽然 f-string 中有条件判断 if ... else ''，但 Python 在编译时会先评估所有表达式，所以即使条件为 False，num_spec_tokens 也必须存在。

修复建议：

# 在使用前先定义 num_spec_tokens
num_spec_tokens = getattr(
    self.speculative_config, "num_speculative_tokens", 0
) if self.speculative_config is not None and self.speculative_config.method is not None else 0

if self.speculative_config is not None and self.speculative_config.method is not None:
    num_spec_tokens = self.speculative_config.num_speculative_tokens
    auto_dispatch_tokens = self.scheduler_config.max_num_seqs * (num_spec_tokens + 1)
else:
    auto_dispatch_tokens = self.scheduler_config.max_num_seqs

或者在日志中使用更简单的表达式：

logger.info(
    f"Auto-setting num_max_dispatch_tokens_per_rank from "
    f"{self.model_config.num_max_dispatch_tokens_per_rank} to {auto_dispatch_tokens} "
    f"(max_num_seqs={self.scheduler_config.max_num_seqs}"
    f"{', num_speculative_tokens=' + str(self.speculative_config.num_speculative_tokens) if self.speculative_config is not None and self.speculative_config.method is not None else ''})."
)