search

Atome-FE / llama-node

Believe in AI democratization. llama for nodejs backed by llama-rs, llama.cpp and rwkv.cpp, work locally on your laptop CPU. support llama/alpaca/gpt4all/vicuna/rwkv model.
https://llama-node.vercel.app/
Apache License 2.0
862 stars 62 forks source link

getting error after second run #71

Closed ralyodio closed 1 year ago

ralyodio commented 1 year ago 
thread 'tokio-runtime-worker' panicked at 'Builder::init should not be called after logger initialized: SetLoggerError(())', /home/runner/.cargo/registry/src/github.com-1ecc6299db9ec823/env_logger-0.10.0/src/lib.rs:816:14stack backtrace:   0: rust_begin_unwind
             at /rustc/84c898d65adf2f39a5a98507f1fe0ce10a2b8dbc/library/std/src/panicking.rs:579:5
   1: core::panicking::panic_fmt
             at /rustc/84c898d65adf2f39a5a98507f1fe0ce10a2b8dbc/library/core/src/panicking.rs:64:14
   2: core::result::unwrap_failed
             at /rustc/84c898d65adf2f39a5a98507f1fe0ce10a2b8dbc/library/core/src/result.rs:1750:5
   3: tokio::loom::std::unsafe_cell::UnsafeCell<T>::with_mut
   4: tokio::runtime::task::raw::poll
   5: tokio::runtime::scheduler::multi_thread::worker::Context::run_task
   6: tokio::runtime::task::raw::poll
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.