| .. |
|
__init__.py
|
04b53d2db5
chore: add initializer files
|
1 gadu atpakaļ |
|
cache_engine.py
|
fe21123a1c
feat: TPU support (#570)
|
7 mēneši atpakaļ |
|
cpu_model_runner.py
|
fdabb55a4d
fix: wrong multi_modal_input format for CPU
|
7 mēneši atpakaļ |
|
cpu_worker.py
|
50b7c13db0
refactor: attention selector (#552)
|
7 mēneši atpakaļ |
|
embedding_model_runner.py
|
8d77c69cbd
feat: support image processor and add llava example
|
7 mēneši atpakaļ |
|
model_runner.py
|
34b41e0a87
chore: add coordinator to reduce code duplication in tp and pp
|
7 mēneši atpakaļ |
|
neuron_model_runner.py
|
35ae01d7ba
refactor: attention metadata term
|
8 mēneši atpakaļ |
|
neuron_worker.py
|
fca911ee0a
vLLM Upstream Sync (#526)
|
8 mēneši atpakaļ |
|
tpu_model_runner.py
|
fe21123a1c
feat: TPU support (#570)
|
7 mēneši atpakaļ |
|
tpu_worker.py
|
a524667db0
fix: device assertion for sdpa backend; fix env for tpu worker
|
7 mēneši atpakaļ |
|
worker.py
|
d0cca80b8b
feat: support sharded tensorizer models
|
7 mēneši atpakaļ |
|
worker_base.py
|
7194047318
remove vllm-nccl
|
7 mēneši atpakaļ |
AlpinDale
34b41e0a87
chore: add coordinator to reduce code duplication in tp and pp