Closed provokateurin closed 11 months ago
Right now only one model is tested in the end-to-end summarize test, but ideally all models should be tested. It is possible that future changes could break one of the models but not the other and this way we'd detect it instantly.
Right now only one model is tested in the end-to-end summarize test, but ideally all models should be tested. It is possible that future changes could break one of the models but not the other and this way we'd detect it instantly.