Qwen/Qwen3-14B
Full fine-tune vs Qwen/Qwen3-14B-Base. 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
199 derivatives checked
- Qwen/Qwen3-14B-AWQ: Quantization
- nvidia/Qwen3-14B-NVFP4: Quantization
- MaziyarPanahi/Qwen3-14B-GGUF: Same weights, converted
- Qwen/Qwen3-14B-GGUF: Quantization
- unsloth/Qwen3-14B-unsloth-bnb-4bit: Quantization
- lmstudio-community/Qwen3-14B-MLX-4bit: Quantization
- lmstudio-community/Qwen3-14B-MLX-8bit: Quantization
- unsloth/Qwen3-14B-GGUF: Same weights, converted
- Qwen/Qwen3-14B-FP8: Quantization
- lmstudio-community/Qwen3-14B-GGUF: Quantization
- richardyoung/Qwen3-14B-abliterated-GGUF: Abliterated
- nvidia/Qwen3-14B-FP8: Quantization
- bartowski/Qwen_Qwen3-14B-GGUF: Same weights, converted
- ggml-org/Qwen3-14B-GGUF: Same weights, converted
- unsloth/Qwen3-14B-bnb-4bit: Quantization
- RedHatAI/Qwen3-14B-NVFP4: Quantization
- unsloth/Qwen3-14B: Same weights, repacked
- cyanelis/ElisNovel-V2-14B: LoRA-style
- mlx-community/Qwen3-14B-4bit: Quantization
- reaperdoesntknow/Symiotic-14B: Same weights, converted
- unsloth/Qwen3-14B-128K-GGUF: Same weights, converted
- reaperdoesntknow/Qemma-Q14B: Different architecture
- Mungert/Hermes-4-14B-GGUF: Full fine-tune
- huihui-ai/Huihui-Qwen3-14B-abliterated-v2: Abliterated
- Aldaris/Qwen3-14B-Q4_K_M-GGUF: Quantization
- geoffmunn/Qwen3-14B-f16: Not comparable yet
- NousResearch/Hermes-4-14B-FP8: Quantization
- NousResearch/Hermes-4-14B: Full fine-tune
- AdarshSingh7647/Eklav-14B-Reranker: LoRA-style
- AdarshSingh7647/Eklav-14B-Math-CotGen: LoRA-style
23 other repos hold byte-for-byte identical weight files.