[torch.compile] E2E correctness testing for fusions
#39.428 aperta il 9 apr 2026
Metriche repository
- Star
- (80.034 star)
- Metriche merge PR
- (Merge medio 9g 2h) (921 PR mergiate in 30 g)
Descrizione
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.