The grimjim/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge galaxy
33 models are built on grimjim/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge: 25 quantized, 0 fine-tuned, 1 adapters and 7 merges. Together they were downloaded 15,695 times in the last 30 days.
Most downloaded direct derivatives
- QuantFactory/llama-3-Nephilim-v3-8B-GGUF (merge), 703 downloads in 30 days
- mradermacher/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge-i1-GGUF (quantized), 414 downloads in 30 days
- grimjim/llama-3-Nephilim-v3-8B (merge), 230 downloads in 30 days
- mradermacher/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge-GGUF (quantized), 128 downloads in 30 days
- featherless-ai-quants/grimjim-Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge-GGUF (quantized), 82 downloads in 30 days
- tensorblock/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge-GGUF (quantized), 60 downloads in 30 days
- grimjim/llama-3-Nephilim-v2-8B (merge), 52 downloads in 30 days
- grimjim/llama-3-Nephilim-v2.1-8B (merge), 52 downloads in 30 days
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.