Describe the bug:
1 -
- Go to 'Models'
- Click on 'Add model'
- App freeze for 10~~ sec
- App closing
Log: nothing
2 -
- Go to 'Model browser'
- Click on 'Any model on trending page'
- App freeze for 10~~ sec
- App closing
Log:
WARN gguf_parser at=src\hf_browser\mod.rs:932:17 | pass2 skip_value failed: kv #29 key=***'tokenizer.ggml.tokens' type=9 pos=524286 remaining=2 ERROR lettuceai_lib::llama_cpp::desktop::engine at=src\infra\utils.rs:551:5 | line 468: Failed to initialize llama backend: BackendAlreadyInitialized
- OS: windows 10, 2.1.1-vulkan app for windows
Expected behavior
Load my GGUF and start chat.
"LettuceAI can run AI models directly on your device using the built-in llama.cpp engine. Browse and download GGUF models from HuggingFace, load them locally, and chat completely offline."
Additional context
I usually use koboldcpp-nocuda and it works perfect.
My app path on my SSD C:\AI\lettuceai
Describe the bug:
1 -
Log: nothing
2 -
Log:
WARN gguf_parser at=src\hf_browser\mod.rs:932:17 | pass2 skip_value failed: kv #29 key=***'tokenizer.ggml.tokens' type=9 pos=524286 remaining=2 ERROR lettuceai_lib::llama_cpp::desktop::engine at=src\infra\utils.rs:551:5 | line 468: Failed to initialize llama backend: BackendAlreadyInitializedExpected behavior
Load my GGUF and start chat.
"LettuceAI can run AI models directly on your device using the built-in llama.cpp engine. Browse and download GGUF models from HuggingFace, load them locally, and chat completely offline."
Additional context
I usually use koboldcpp-nocuda and it works perfect.
My app path on my SSD C:\AI\lettuceai