To run this implementation, the nightly version of triton and torch is going to be mounted. This Edition can be operate on a single 80GB GPU for gpt-oss-120b. anyways , I'm happy which i was in the position to help you in no matter what ways in which my ideas https://brookslkdjg.ampblogs.com/helping-the-others-realize-the-advantages-of-hbs-case-study-solution-74005009