AXCXEPT/Qwen3-EZO-8B-beta
Full fine-tune vs Qwen/Qwen3-8B. Fine-tune: 92% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 0/1, Norms 95/145, MLP 108/108, Attention 144/144, lm_head 1/1
17 derivatives checked
- mradermacher/Qwen3-EZO-8B-beta-i1-GGUF: Quantization
- mmnga/Qwen3-EZO-8B-beta-gguf: Quantization
- DevQuasar/AXCXEPT.Qwen3-EZO-8B-beta-GGUF: Same weights, converted
- mradermacher/Qwen3-EZO-8B-beta-GGUF: Same weights, converted
- ryanfortin/community-blend-qwen3-8b: Full fine-tune
- ECE-ILAB/POIROT-ECE-1.2: Full fine-tune
- ZeroXClem/Qwen3-8B-HoneyBadger-EXP: Full fine-tune
- YOYO-AI/Qwen3-8B-YOYO-V2-Hybrid: Full fine-tune
- YOYO-AI/Qwen3-EZO-8B-YOYO-slerp-128K: Full fine-tune
- YOYO-AI/Qwen3-EZO-8B-YOYO-nuslerp-plus-128K: Full fine-tune
- YOYO-AI/Qwen3-EZO-8B-YOYO-slerp: Full fine-tune
- ECE-ILAB/POIROT-ECE-1.0: Partial fine-tune
- YOYO-AI/Qwen3-EZO-8B-YOYO-karcher-128K: Full fine-tune
- YOYO-AI/Qwen3-EZO-8B-YOYO-nuslerp-plus: Full fine-tune
- YOYO-AI/Qwen3-EZO-8B-YOYO-nuslerp: Full fine-tune
- ECE-ILAB/POIROT-ECE-1.1: Full fine-tune
- YOYO-AI/Qwen3-EZO-8B-YOYO-nuslerp-128K: Full fine-tune