Reiterate3680's picture
Update README.md
00f1e1e verified
metadata
base_model: Nitral-AI/Sekhmet_Bet-L3.1-8B-v0.2
language:
  - en
license: other
pipeline_tag: text-generation
quantized_by: Reiterate3680

Fixed 3.1 GGUFs require KoboldCPP 1.17.1 or newer to run.

Unfixed: https://huggingface.co/Reiterate3680/Sekhmet_Bet-L3.1-8B-v0.2-GGUF-BAD-LONG-CONTEXT

Original Model: https://huggingface.co/Nitral-AI/Sekhmet_Bet-L3.1-8B-v0.2

made with https://huggingface.co/FantasiaFoundry/GGUF-Quantization-Script

Models Q2_K_L, Q4_K_L, Q5_K_L, Q6_K_L, are using Q_8 output tensors and token embeddings

using bartowski's imatrix dataset