[torch.compile] E2E correctness testing for fusions
#39.428 geöffnet am 9. Apr. 2026
Repository-Metriken
- Stars
- (80.034 Stars)
- PR-Merge-Metriken
- (Durchschn. Merge 9T 2h) (921 gemergte PRs in 30 T)
Beschreibung
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.