NousResearch/Hermes-4-14B
Full fine-tune vs Qwen/Qwen3-14B. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 1/1, Norms 158/161, MLP 120/120, Attention 160/160, lm_head 1/1
21 derivatives checked
- bartowski/NousResearch_Hermes-4-14B-GGUF: Same weights, converted
- gabriellarson/Hermes-4-14B-GGUF: Same weights, converted
- mradermacher/Hermes-4-14B-i1-GGUF: Quantization
- mlx-community/Hermes-4-14B-4bit: Quantization
- mradermacher/Hermes-4-14B-GGUF: Quantization
- genevaquant/NousResearch_Hermes-4-14B-GGUF: Same weights, converted
- cyankiwi/Hermes-4-14B-AWQ-4bit: Quantization
- DevQuasar/NousResearch.Hermes-4-14B-GGUF: Same weights, converted
- NikolayKozloff/Hermes-4-14B-Q4_K_M-GGUF: Quantization
- mlx-community/NousResearch_Hermes-4-14B-BF16-abliterated-mlx: LoRA-style
- fasolack/Hermes-4-14B-W4A16: Quantization
- mlx-community/Hermes-4-14B-8bit: Quantization
- NikolayKozloff/Hermes-4-14B-Q5_K_M-GGUF: Quantization
- warshanks/Hermes-4-14B-AWQ: Quantization
- tonythethompson/Hermes-4-14B-NVFP4: Quantization
- SandLogicTechnologies/Hermes-4-14B-GGUF: Quantization
- mlx-community/Hermes-4-14B-6bit: Quantization
- mhaxscp/Hermes4-Philosopher-Agent: Full fine-tune
- RavichandranJ/Hermes-4-14B-OBLITERATED: Full fine-tune
- jbrashear/Aegis-14B: LoRA-style
- RavichandranJ/Hermes-4-14B-OBLITERATED-V2: Full fine-tune
2 other repos hold byte-for-byte identical weight files.