AlpinDale 5b0c11d190 support pipeline parallel pynccl groups 6 месяцев назад
..
__init__.py 04b53d2db5 chore: add initializer files 1 год назад
cache_engine.py 50b7c13db0 refactor: attention selector (#552) 6 месяцев назад
cpu_model_runner.py f6250c5516 move dockerfiles to root; fix cpu build 6 месяцев назад
cpu_worker.py 50b7c13db0 refactor: attention selector (#552) 6 месяцев назад
embedding_model_runner.py de62ceb18c refactor: eliminate parallel worker per-step task scheduling overhead 6 месяцев назад
model_runner.py 5b0c11d190 support pipeline parallel pynccl groups 6 месяцев назад
neuron_model_runner.py 35ae01d7ba refactor: attention metadata term 6 месяцев назад
neuron_worker.py fca911ee0a vLLM Upstream Sync (#526) 7 месяцев назад
worker.py eb2c5c77df feat: enforce the max possible seqlen 6 месяцев назад
worker_base.py de62ceb18c refactor: eliminate parallel worker per-step task scheduling overhead 6 месяцев назад