vllm.worker
Modules:
Name | Description |
---|---|
cache_engine |
CacheEngine class for managing the KV cache. |
cpu_enc_dec_model_runner |
|
cpu_model_runner |
|
cpu_pooling_model_runner |
|
cpu_worker |
A CPU worker class. |
enc_dec_model_runner |
|
hpu_model_runner |
|
hpu_worker |
|
model_runner |
|
model_runner_base |
|
multi_step_hpu_worker |
|
multi_step_model_runner |
|
multi_step_neuron_model_runner |
|
multi_step_neuronx_distributed_model_runner |
|
multi_step_tpu_worker |
|
multi_step_worker |
|
neuron_model_runner |
|
neuron_worker |
A Neuron worker class. |
neuronx_distributed_model_runner |
|
pooling_model_runner |
|
tpu_model_runner |
|
tpu_worker |
|
utils |
Worker-related helper functions. |
worker |
A GPU worker class. |
worker_base |
|
xpu_model_runner |
|
xpu_worker |
A XPU worker class. |