diff --git a/README.md b/README.md index 45c6d2c1d..27f06685e 100644 --- a/README.md +++ b/README.md @@ -34,6 +34,9 @@ After running this command you can launch Koboldcpp from the current directory u ## Run on RunPod - KoboldCpp can now be used on RunPod cloud GPUs! This is an easy way to get started without installing anything in a minute or two, and is very scalable, capable of running 70B+ models at afforable cost. [Try our RunPod image here!](https://koboldai.org/runpodcpp). +## Run on Novita AI +KoboldCpp can now also be run on Novita AI, a newer alternative GPU cloud provider which has a quick launch KoboldCpp template for as well. [Check it out here!](https://koboldai.org/novitacpp) + ## Docker - The official docker can be found at https://hub.docker.com/r/koboldai/koboldcpp - If you're building your own docker, remember to set CUDA_DOCKER_ARCH or enable LLAMA_PORTABLE diff --git a/expose.cpp b/expose.cpp index 7fec5cea9..b8b469eca 100644 --- a/expose.cpp +++ b/expose.cpp @@ -78,7 +78,7 @@ extern "C" if(file_format==FileFormat::GPTJ_1 || file_format==FileFormat::GPTJ_2 || file_format==FileFormat::GPTJ_3 || file_format==FileFormat::GPTJ_4 || file_format==FileFormat::GPTJ_5) { - printf("\n---\nIdentified as GPT-J model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nIdentified as Legacy GPT-J model: (ver %d)\nAttempting to Load...\n---\n", file_format); ModelLoadResult lr = gpttype_load_model(inputs, file_format, file_format_meta); if (lr == ModelLoadResult::RETRY_LOAD) { @@ -87,14 +87,14 @@ extern "C" //if we tried 1 first, then try 3 and lastly 2 //otherwise if we tried 3 first, then try 2 file_format = FileFormat::GPTJ_4; - printf("\n---\nRetrying as GPT-J model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nRetrying as Legacy GPT-J model: (ver %d)\nAttempting to Load...\n---\n", file_format); lr = gpttype_load_model(inputs, file_format, file_format_meta); } if (lr == ModelLoadResult::RETRY_LOAD) { file_format = FileFormat::GPTJ_3; - printf("\n---\nRetrying as GPT-J model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nRetrying as Legacy GPT-J model: (ver %d)\nAttempting to Load...\n---\n", file_format); lr = gpttype_load_model(inputs, file_format, file_format_meta); } @@ -102,7 +102,7 @@ extern "C" if (lr == ModelLoadResult::RETRY_LOAD) { file_format = FileFormat::GPTJ_2; - printf("\n---\nRetrying as GPT-J model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nRetrying as Legacy GPT-J model: (ver %d)\nAttempting to Load...\n---\n", file_format); lr = gpttype_load_model(inputs, file_format, file_format_meta); } } @@ -118,18 +118,18 @@ extern "C" } else if(file_format==FileFormat::GPT2_1||file_format==FileFormat::GPT2_2||file_format==FileFormat::GPT2_3||file_format==FileFormat::GPT2_4) { - printf("\n---\nIdentified as GPT-2 model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nIdentified as Legacy GPT-2 model: (ver %d)\nAttempting to Load...\n---\n", file_format); ModelLoadResult lr = gpttype_load_model(inputs, file_format, file_format_meta); if (lr == ModelLoadResult::RETRY_LOAD) { file_format = FileFormat::GPT2_3; - printf("\n---\nRetrying as GPT-2 model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nRetrying as Legacy GPT-2 model: (ver %d)\nAttempting to Load...\n---\n", file_format); lr = gpttype_load_model(inputs, file_format, file_format_meta); } if (lr == ModelLoadResult::RETRY_LOAD) { file_format = FileFormat::GPT2_2; - printf("\n---\nRetrying as GPT-2 model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nRetrying as Legacy GPT-2 model: (ver %d)\nAttempting to Load...\n---\n", file_format); lr = gpttype_load_model(inputs, file_format, file_format_meta); } if (lr == ModelLoadResult::FAIL || lr == ModelLoadResult::RETRY_LOAD) @@ -143,27 +143,27 @@ extern "C" } else if(file_format==FileFormat::NEOX_1 || file_format==FileFormat::NEOX_2 || file_format==FileFormat::NEOX_3 || file_format==FileFormat::NEOX_4 || file_format==FileFormat::NEOX_5 || file_format==FileFormat::NEOX_6 || file_format==FileFormat::NEOX_7) { - printf("\n---\nIdentified as GPT-NEO-X model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nIdentified as Legacy GPT-NEO-X model: (ver %d)\nAttempting to Load...\n---\n", file_format); ModelLoadResult lr = gpttype_load_model(inputs, file_format, file_format_meta); if (lr == ModelLoadResult::RETRY_LOAD) { if(file_format==FileFormat::NEOX_2) { file_format = FileFormat::NEOX_3; - printf("\n---\nRetrying as GPT-NEO-X model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nRetrying as Legacy GPT-NEO-X model: (ver %d)\nAttempting to Load...\n---\n", file_format); lr = gpttype_load_model(inputs, file_format, file_format_meta); } else { file_format = FileFormat::NEOX_5; - printf("\n---\nRetrying as GPT-NEO-X model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nRetrying as Legacy GPT-NEO-X model: (ver %d)\nAttempting to Load...\n---\n", file_format); lr = gpttype_load_model(inputs, file_format, file_format_meta); } } if (lr == ModelLoadResult::RETRY_LOAD) { file_format = FileFormat::NEOX_1; - printf("\n---\nRetrying as GPT-NEO-X model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nRetrying as Legacy GPT-NEO-X model: (ver %d)\nAttempting to Load...\n---\n", file_format); lr = gpttype_load_model(inputs, file_format, file_format_meta); } if (lr == ModelLoadResult::FAIL || lr == ModelLoadResult::RETRY_LOAD) @@ -179,16 +179,20 @@ extern "C" { if(file_format==FileFormat::MPT_1) { - printf("\n---\nIdentified as MPT model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nIdentified as Legacy MPT model: (ver %d)\nAttempting to Load...\n---\n", file_format); } else if(file_format==FileFormat::RWKV_1 || file_format==FileFormat::RWKV_2) { - printf("\n---\nIdentified as RWKV model: (ver %d)\nAttempting to Load...\n---\n", file_format); + printf("\n---\nIdentified as Legacy RWKV model: (ver %d)\nAttempting to Load...\n---\n", file_format); } else if(file_format==FileFormat::GGUF_GENERIC) { printf("\n---\nIdentified as GGUF model: (ver %d)\nAttempting to Load...\n---\n", file_format); } + else if(file_format==FileFormat::GGML || file_format==FileFormat::GGHF || file_format==FileFormat::GGJT || file_format==FileFormat::GGJT_2 || file_format==FileFormat::GGJT_3) + { + printf("\n---\nIdentified as Legacy GGML model: (ver %d)\nYou are STRONGLY ENCOURAGED to obtain a newer GGUF model!\nAttempting to Load...\n---\n", file_format); + } else { printf("\n---\nUnidentified Model Encountered: (ver %d)\n---\n", file_format);