help-forum

805 threads · Page 10 of 17

What can be considered as 'best' model for instruction-chat in ooba with a 8GB card? 37 messages
Windows
promt format for web generation ui model 3 messages
training have been trying to train vicuna 13b to speak hinglish..it ends up speaking gibberish 5 messages
ERROR: Could not build wheels for llama-cpp-python 33 messages
Setup LLaMA
xformers and model cohesion degradation... 8 messages
Fine-tuning quantized Llama2 4 messages
exllamav2 loading attempts - WSL2 on W11 266 messages
Linux
How to load gguf format 3 messages
Cannot install with the Windows Installer - OpenSSL error? 2 messages
Setup Windows
Problems when running GPTQ and GGUF models 10 messages
LLaMA Windows
Error starting .bat 9 messages
Windows
oobabot-plugin 16 messages
Running a larger LLM model on mix of vRAM (GPU) + RAM - possible? 3 messages
Can't load any GGUF model 25 messages
Windows
CUDA Out Of Memory with 0 allocated? 5 messages
Windows
ModuleNotFoundError: No module named 'gptq_for_llama' 5 messages
Linux LLaMA
Llama-2 LoRA fine-tuning based on books/novels help 10 messages
prompt template when streaming via port 5005 2 messages
Model Reply Blank 27 messages
I am getting errors about no path existing, no model being loaded and an unlocatable checkpoint file 15 messages
From origin has been blocked by CORS policy 2 messages
Notebook Tab - Text generation stops before 2k context is reached 2 messages
Webui api returns an error while running on second gpu or split. 23 messages
Linux Setup Hardware
Can't generate using ExLlama... 2 messages
Linux Setup Hardware LLaMA
Problems with autoGPTQ 3 messages
Linux
gguf loads slower than ggml 30 messages
Linux LLaMA Hardware
unrecognized arguments 19 messages
[solved] error "Illegal instruction (core dumped)" when trying to use llama.cpp. 7 messages
Linux LLaMA AMD Windows
Absence of ExLlama. 30 messages
Linux LLaMA AMD
Mac M1 Air issue 53 messages
unexpected keyword argument 'mul_mat_q' 5 messages
Gradio Linux Setup
Which model for my usage ? 3 messages
Hardware Setup Windows
cuda out of memory 36 messages
Linux LLaMA Hardware
RuntimeError: The installed version of g++ (12.3.0) is greater than the maximum req by CUDA 11.7 3 messages
Linux
Can execute the start_windows.bat! There is no web ui. No nothing. Only an empty shell 28 messages
Chat edit 2 messages
"Continue" only adds a single word 5 messages
Cant launch any models I get an error that llama.cpp is not installed. How do I install it? 136 messages
Unable to load ggml-llama-q4_0.bin 10 messages
Windows
Accessing web UI API through Jupyter notebook on Runpod.io 3 messages
Hardware LLaMA
Just got new error (traceback error) when trying to load model 7 messages
Using default generate UI seems to create different output than the api even using same preset 7 messages
The model is loaded without errors but no replies! 9 messages
Hardware Setup Prompts & Characters Windows
best settings for GGML models (13b), 6gb vram, 16gb ram for best speed 92 messages
Windows
Download HF Model Branch 3 messages
I am a complete donut and am trying to figure everything out, can I get someone to help me? 3 messages
Download Llama 2 hugging face 243 messages
Linux Setup LLaMA
Getting error in Colab 3 messages
Am kind of lost even after reading the instructions 4 messages
I get this error when trying to launch the start_windows.bat file 70 messages