vllm-project/vllm

[torch.compile] E2E correctness testing for fusions

Open

#39,428 opened on 2026年4月9日

GitHub で見る
 (6 comments) (0 reactions) (0 assignees)Python (80,034 stars) (16,816 forks)batch import
help wantedtorch.compile

説明

E2E tests for fusions (tests/compile/fusions_e2e) have done a great job preventing fusion regressions where model/forward code changes break a custom torch.compile fusion pass. However, we currently have no way of testing correctness for these fusion configurations.

It would be good to investigate an approach where we only run a few layers of a model and compare the outputs. This would be helpful for correctness testing in general, and we could compare the outputs to both a baseline vLLM configuration and the huggingface baseline.

This would likely require some work to fix weight loading for models like DeepSeek when --hf-overrides.num_hidden_layers is overriden.

コントリビューターガイド

技術スタック
pythonpytorch
領域
backendtesting
Issue 種別
feature
難度新しい貢献者にとっての実装難度です。1 は非常に小さな変更、5 は専門的な作業です。
4
推定時間経験ある貢献者が調査、実装、テスト、pull request 準備にかけるおおよその時間範囲です。
3-5 days
活動状況Issue が今どれだけ取り組みやすいかを示します。新しい、活発、古い、ブロック中、またはメンテナー入力待ちなどです。
active
明確さ期待される変更、受け入れ条件、次の手順がどれだけ明確かを示します。
needs investigation
前提条件
PyTorch compilationvLLM fusion passesModel weight loading
初心者向け度初回貢献者にどれだけ取り組みやすいかを 1-100 で推定したスコアです。
35
調査方針
Start by examining the existing test suite at `tests/compile/fusions e2e` to understand the structure. Consider using a subset of layers with weight loading overrides via ` hf overrides.num hidden layers`, especially for models like DeepSeek. Develop a method to compare outputs against both vLLM baseline and Hugging Face reference, possibly using a small model configuration. Address any weight loading issues that arise from overriding the number of layers.