metadata
license: other
Star-Command-R-32B-v1 - EXL2 4.5bpw
This is a 4.5bpw EXL2 quant of TheDrummer/Star-Command-R-32B-v1
This quant was made using exllamav2-0.2.0 with default dataset.
I tested this quant shortly in some random RPs (including 8k+ RPs where remembering and understanding specific facts in the context is needed) and it seems to work fine. In my short tests it seemed better than GGUF of similar size (Q4_K_M).
This quant fits nicely in 24GB VRAM, especially with Q4 cache.
Prompt Templates
Uses Command-R format.
Original readme below
Join our Discord! https://discord.gg/Nbv9pQ88Xb
BeaverAI proudly presents...
Star Command R 32B v1 🌟
An RP finetune of Command-R-8-2024
Links
- Original: https://huggingface.co/TheDrummer/Star-Command-R-32B-v1
- GGUF: https://huggingface.co/TheDrummer/Star-Command-R-32B-v1-GGUF
Usage
- Cohere Instruct format or Text Completion
Special Thanks
- Mr. Gargle for the GPUs! Love you, brotha.