| .. |
|
__init__.py
|
04b53d2db5
chore: add initializer files
|
há 1 ano atrás |
|
cache_engine.py
|
bf88c8567e
feat: mamba model support (#674)
|
há 4 meses atrás |
|
cpu_model_runner.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
cpu_worker.py
|
22a4cd4595
core: fix spec decode metrics and envs circular import (#889)
|
há 1 mês atrás |
|
embedding_model_runner.py
|
89a2c6dee1
chore: refactor `MultiModalConfig` initialization and profiling (#745)
|
há 4 meses atrás |
|
enc_dec_model_runner.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
model_runner.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
model_runner_base.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
multi_step_model_runner.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
multi_step_worker.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
neuron_model_runner.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
neuron_worker.py
|
0c6d90dade
neuron: add support for tensor parallelism (#923)
|
há 1 mês atrás |
|
openvino_model_runner.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
openvino_worker.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
tpu_model_runner.py
|
5bec8fbb1b
tpu: add support for async postprocessing (#968)
|
há 1 mês atrás |
|
tpu_worker.py
|
ea59784f59
tpu: remove torch._dynamo.reset() (#952)
|
há 1 mês atrás |
|
utils.py
|
b3f6eeb1d2
vlm: increase the default `max_num_batched_tokens` for multimodal models (#973)
|
há 1 mês atrás |
|
worker.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
worker_base.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
xpu_model_runner.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
há 1 mês atrás |
|
xpu_worker.py
|
15cb8d5c26
xpu: support pipeline parallel (#932)
|
há 1 mês atrás |
AlpinDale
b3f6eeb1d2
vlm: increase the default `max_num_batched_tokens` for multimodal models (#973)