Yang

I finish the first step, but find the second step make SharpieBinder is not possible to finish. There is no source code for libclang-mono, and I don't know whether the...

I use Next.js with tauri, so need this feature. It will b e really nice otherwise have to downgrade to a lower version. :-(

I use ubuntu. Cuda 11.8 and the error message is below: ``` Loaded hyperparameters ggml ctx size = 0.09 MB ggml_init_cublas: found 1 CUDA devices: Device 0: NVIDIA GeForce RTX...

I am not sure it is the reason since I tested on the newest version llama.cpp. When I directly use llama.cpp,it has no issue at all: ``` llama_model_load_internal: using CUDA...

I checked the flag in build.rs for ggml-sys and try to make them the same as below but it still report the same error. I also tried the branch you...

Sure, I will try and get back to you once I get the result.

@LLukas22 I checkout the new llm code and run the following code: cargo run --release --features cublas --example vicuna-chat llama orca-mini-13b.ggmlv3.q4_0.bin And it reports the following error: Loaded hyperparameters ggml...

@LLukas22 Sure, it also bugs me. My NVIDIA Dirver Version: 520.61.05, CUDA Version: 11.8