Quantization made by Richard Erkhov.

Meta-Llama-3.1-8B-Instruct-ablated-v1 - bnb 8bits

Llama 3.1 instruct 8b 128k context with ablation.

This means it will refuse less of your valid requests for an uncensored UX. This version is an improvement on the v0. Use this 8b for best results.

Use responsibly and at your own discretion. We do not take any responsibility for how you apply this intelligence, just as we do not for how you apply your own.