search

VITA-MLLM / VITA

✨✨VITA: Towards Open-Source Interactive Omni Multimodal LLM
Other
871 stars 46 forks source link

The number of required GPUs exceeds the total number of available GPUs in the placement group. #42

Open JosenJin opened 3 weeks ago

JosenJin commented 3 weeks ago

2024-09-21 18:15:28,304 INFO worker.py:1786 -- Started a local Ray instance. 2024-09-21 18:15:28,399 INFO worker.py:1786 -- Started a local Ray instance. Process Process-2: Traceback (most recent call last): File "D:\ai\anaconda3\envs\vita\Lib\multiprocessing\process.py", line 314, in _bootstrap self.run() File "D:\ai\anaconda3\envs\vita\Lib\multiprocessing\process.py", line 108, in run self._target(*self._args, *self._kwargs) File "D:\ai\VITA\web_demo\web_interactive_demo.py", line 127, in load_model llm = AsyncLLMEngine.from_engine_args(engine_args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ai\anaconda3\envs\vita\Lib\site-packages\vllm\engine\async_llm_engine.py", line 570, in from_engine_args executor_class = cls._get_executor_cls(engine_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ai\anaconda3\envs\vita\Lib\site-packages\vllm\engine\async_llm_engine.py", line 546, in _get_executor_cls initialize_ray_cluster(engine_config.parallel_config) File "D:\ai\anaconda3\envs\vita\Lib\site-packages\vllm\executor\ray_utils.py", line 265, in initialize_ray_cluster raise ValueError( ValueError: The number of required GPUs exceeds the total number of available GPUs in the placement group. Process Process-3: Traceback (most recent call last): File "D:\ai\anaconda3\envs\vita\Lib\multiprocessing\process.py", line 314, in _bootstrap self.run() File "D:\ai\anaconda3\envs\vita\Lib\multiprocessing\process.py", line 108, in run self._target(self._args, **self._kwargs) File "D:\ai\VITA\web_demo\web_interactive_demo.py", line 127, in load_model llm = AsyncLLMEngine.from_engine_args(engine_args) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ai\anaconda3\envs\vita\Lib\site-packages\vllm\engine\async_llm_engine.py", line 570, in from_engine_args executor_class = cls._get_executor_cls(engine_config) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ai\anaconda3\envs\vita\Lib\site-packages\vllm\engine\async_llm_engine.py", line 546, in _get_executor_cls initialize_ray_cluster(engine_config.parallel_config) File "D:\ai\anaconda3\envs\vita\Lib\site-packages\vllm\executor\ray_utils.py", line 265, in initialize_ray_cluster raise ValueError( ValueError: The number of required GPUs exceeds the total number of available GPUs in the placement group.

longzw1997 commented 3 weeks ago

May I ask what your specific configuration is?

JosenJin commented 3 weeks ago

2080ti 22G

longzw1997 commented 3 weeks ago

We haven't tried it on the 2080ti. You can try using 8 * 2080ti 22G, or use GPUs with larger memory.