[BugFix] Fix deep gemm import

[RichardWooSJTU]

Motivation

💡 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

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!

[PaddlePaddle-bot]

🤖 AI Code Review | 2026-04-16

📋 Review 摘要

PR 概述：修复 DeepSeekV3 模型中 deep_gemm 的导入问题，统一使用 fp8_utils 中加载的 deep_gemm 模块。

变更范围：model_executor/models/deepseek_v3.py

影响面 Tag：[Models] [Quantization]

📝 PR 规范检查

PR 描述不完整，建议补充以下内容：

Motivation 建议（可直接复制）：

原代码直接使用 `import deep_gemm` 导入第三方 deep-gemm 包，可能与 FastDeploy 内置的 deep_gemm 实现冲突。修改为从 `fp8_utils` 导入可确保使用 FastDeploy 内置的实现。同时删除了冗余的 `paddle.enable_compat(scope={"deep_gemm"})` 调用，因为 `fp8_utils.load_deep_gemm()` 已根据 SM 版本正确处理兼容性设置。

Modifications 建议（可直接复制）：

1. 删除模块级别的 `paddle.enable_compat(scope={"deep_gemm"})` 调用
2. 将 forward() 函数内的 `import deep_gemm` 修改为 `from fastdeploy.model_executor.layers.quantization.fp8_utils import deep_gemm`

Checklist 建议：

• Add at least a tag in the PR title.
• Format your code, run pre-commit before commit.
• Add unit tests. (建议补充说明：此 PR 主要是导入路径修正，无需新增测试)
• Provide accuracy results.
• N/A（未提交到 release 分支）

问题

未发现阻塞性问题。

总体评价

这是一个合理的 BugFix 修复。修改正确地统一了 deep_gemm 的导入方式，避免直接导入第三方 deep-gemm 包可能导致的冲突。删除冗余的 enable_compat 调用也是正确的，因为 fp8_utils 模块在加载 deep_gemm 时已根据硬件架构正确处理了兼容性设置。建议完善 PR 描述后合并。

[codecov-commenter]

Codecov Report

❌ Patch coverage is 0% with 1 line in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@e53f518). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/model_executor/models/deepseek_v3.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #7425   +/-   ##
==========================================
  Coverage           ?   74.15%           
==========================================
  Files              ?      394           
  Lines              ?    54760           
  Branches           ?     8578           
==========================================
  Hits               ?    40610           
  Misses             ?    11411           
  Partials           ?     2739           

Flag	Coverage Δ
GPU	74.15% <0.00%> (?)

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.