Open liuyang77886 opened 1 year ago
Have the same issue with ggml-gpt4all-j.bin and
2023-06-22 01:22:51 8:22AM DBG Loading model 'ggml-gpt4all-j.bin' greedly 2023-06-22 01:22:51 8:22AM DBG [llama] Attempting to load 2023-06-22 01:22:51 8:22AM DBG Loading model llama from ggml-gpt4all-j.bin 2023-06-22 01:22:51 8:22AM DBG Loading model in memory from file: /models/ggml-gpt4all-j.bin 2023-06-22 01:22:51 ggml_init_cublas: found 1 CUDA devices: 2023-06-22 01:22:51 Device 0: NVIDIA GeForce RTX 4090 2023-06-22 01:22:52 llama.cpp: loading model from /models/ggml-gpt4all-j.bin 2023-06-22 01:22:56 error loading model: unexpectedly reached end of file 2023-06-22 01:22:56 llama_init_from_file: failed to load model 2023-06-22 01:22:56 fatal error: unexpected signal during runtime execution 2023-06-22 01:22:56 [signal SIGSEGV: segmentation violation code=0x1 addr=0x4 pc=0xa82190]
Found the reason candidate. LocalAI tries to load model with different bindings. Once it fails it moves to the next binding candidate. However, this method cannot be applied with CUDA because CUDA throws Segment Fault. This terminates Docker container as SIG* kernel command was generated from the application as designed in Docker.
Having the same issue. @yunghoy did you find a workaround ?
Hi! I'm a bot running with LocalAI ( a crazy experiment of @mudler ) - please beware that I might hallucinate sometimes!
_but.... I can also be funny or helpful :smilecat: and I can provide generally speaking good tips or places where to look after in the documentation or in the code based on what you wrote in the issue.
Don't engage in conversation with me, I don't support (yet) replying!
The issue seems to be related to a segmentation fault in the LocalAI application. It could be caused by a number of reasons such as a programming error, memory allocation issues or a hardware failure. To help debug the issue further, you can try to reproduce the problem and gather more information about the environment and the code that is being executed at the time of the crash. You can also check the logs for any error messages or stack traces that may provide additional clues. Additionally, you can try to reproduce the issue on a different machine or platform to rule out any hardware-related issues.
Sources:
环境信息(Environment information):
不论我调用ggml-gpt4all-j 或者 ggml-gpt4all-j-v1.3-groovy.bin 都报错(I got errors with models below):