A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 03:00 UTC
[Bugfix] Update neuron_executor.py to add optional vision_language_config (#3695)
Adam Boeglin committed
Mar 28, 2024 at 10:43 UTC
1715056fef0e8f047f71bfb864fe9773af47af41
1 file changed
+2
-1
vllm/executor/neuron_executor.py
+2
-1
@@ -1,7 +1,7 @@
1
from typing import Dict, List, Optional
2
3
from vllm.config import (CacheConfig, DeviceConfig, LoRAConfig, ModelConfig,
4
- ParallelConfig, SchedulerConfig)
4
+ ParallelConfig, SchedulerConfig, VisionLanguageConfig)
5
from vllm.executor.executor_base import ExecutorBase
6
from vllm.logger import init_logger
7
from vllm.lora.request import LoRARequest
@@ -20,6 +20,7 @@ class NeuronExecutor(ExecutorBase):
20
scheduler_config: SchedulerConfig,
21
device_config: DeviceConfig,
22
lora_config: Optional[LoRAConfig],
23
+ vision_language_config: Optional[VisionLanguageConfig],
24
) -> None:
25
self.model_config = model_config
26
self.cache_config = cache_config