AlpinDale
|
d8c4193704
feat: Speculative Decoding using a draft model (#432)
|
9 месяцев назад |
AlpinDale
|
083ba7b452
roll back chunked prefill changes to SDPA, isolate cpu worker
|
9 месяцев назад |
AlpinDale
|
8c67b37131
fix docstrings
|
9 месяцев назад |
AlpinDale
|
50c2434267
move megatron to a top-level directory
|
9 месяцев назад |
AlpinDale
|
4d33ce60da
feat: Triton flash attention backend for ROCm (#407)
|
9 месяцев назад |
AlpinDale
|
9e9057a614
separate init_distributed_environment from worker
|
9 месяцев назад |
AlpinDale
|
a304f76d89
feat: Intel CPU support (#403)
|
9 месяцев назад |