Neko-Institute-of-Science

/

VicUnLocked-30b-LoRA

like 16

I've been watching this for a while and looking forward to it being ready. Thanks very much for getting it trained!

I've done a merge with LLaMA 30B, and quantisations:

• 4-bit, 5-bit and 8-bit GGML models for CPU (+CUDA) inference.
• 4-bit model GPTQ model for GPU inference.
• float16 HF format model for GPU inference and further conversions.