Training a model without a GPU
Looking for a low-end setup
Mistral shocks AI community as latest open source model eclipses GPT-3.5 performance
Mixture of Experts Explained (Huggingface blog)
Mistral releases version 0.2 of their 7B model
QuIP#: SOTA 2 bit LLMs
ByteDance AI Promises Stronger than Gemini Open Weight GPT Dropping Soon
Mistral drops a new magnet download
Inside the A.I. Arms Race That Changed Silicon Valley Forever
LLMs made into single-file executables with llamafile
Unsloth: 80% faster 50% less memory LLM finetuning
I'm having a fantastic time with this model.
Orca 2: Teaching Small Language Models How to Reason
Hundreds of OpenAI employees threaten to resign and join Microsoft
Catch me if you can! How to beat GPT-4 with a 13B model | LMSYS Org
TensorRT-LLM evaluation of the new H200 GPU achieves 11,819 tokens/s on Llama2-13B
...
ExUI - a lightweight web UI for ExLlamaV2 by turboderp
New "Context Shifting" feature in KoboldCPP 1.48
