Tim Hillison

ep1tim

·

AI & ML interests

None yet

Recent Activity

reacted to mike18's post with 🔥 about 10 hours ago

We've released smaller GGUF builds of Clef, Clef-Flash, Perplexity's Decider and Kev for stock llama.cpp, plus Gemma 4 E4B and E2B. Each file is checked against its original, with the full comparisons on its card. One example: Clef-Flash goes from 18.2 GB to 4.9 GB and gives the original's answer on 95% of 2,000 test decisions on CPU. https://huggingface.co/turintech/Clef-Flash-Artemis-GGUF Which model would you like smaller next, and on what hardware would you run it? Reply here or in our feedback thread: https://huggingface.co/spaces/turintech/README/discussions/1 Decision models: https://huggingface.co/collections/turintech/artemis-edge-decision-models-6accea74eb7016a886f25fdb Compressed LLMs: https://huggingface.co/collections/turintech/artemis-edge-compressed-llms-6acd251457ba2b9f33ad9d8a

View all activity

Organizations

ep1tim 's datasets

None public yet