unsloth/Qwen3-4B
Same weights, repacked vs Qwen/Qwen3-4B. Same weights in different files (re-saved, re-sharded or new metadata).
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 0/108, Attention 0/144
25 derivatives checked
- PrimitivePenguin/brainrot-qwen3-4b-GGUF: Quantization
- rizkyayub/rizbuy-submission-qwen3-indonesian: LoRA-style
- giocorte/totem-slm-rl-v4: LoRA-style
- giocorte/totem-slm-rl-v3: LoRA-style
- lschaffer/qwen3-4b-weathersensorsmcp-ollama: LoRA-style
- lrsbrgrn/HalluGuard-Qwen3-4B: LoRA-style
- ermiaazarkhalili/Qwen3-4B-SFT-Fable5-GGUF: Full fine-tune
- lschaffer/qwen3-4b-weathersensorsmcp: LoRA-style
- khazarai/Fino1-4B: LoRA-style
- TeichAI/Qwen3-4B-GPT-5.2-High-Reasoning-Distill: LoRA-style
- iamdyeus/qwendean-4b: LoRA-style
- mlkro/Qwen3-4B-PT-SynthDolly-1B: LoRA-style
- Sakalti/Mystica1-4B: Same weights, converted
- ermiaazarkhalili/Qwen3-4B-Function-Calling-xLAM-Unsloth: LoRA-style
- patrickleenyc/Qwen3-4B-TwoFace: LoRA-style
- prometheus04/Qwen3-4B-TerminalBench: LoRA-style
- mlkro/Qwen3-4B-EN-SynthDolly-2B: LoRA-style
- mlkro/Qwen3-4B-EN-SynthDolly-3B: LoRA-style
- ermiaazarkhalili/Qwen3-4B-SFT-Claude-Opus-Reasoning-Unsloth: LoRA-style
- mlkro/Qwen3-4B-EN-SynthDolly-1B: LoRA-style
- mlkro/Qwen3-4B-PT-SynthDolly-3B: LoRA-style
- taharmasmaliyev07/Mirza-Qwen3-4B-Editor-E3-BF16-V2: LoRA-style
- OmAlve/reading-steiner-qwen3-4b: LoRA-style
- mlkro/Qwen3-4B-PT-SynthDolly-2B: LoRA-style
- ChavyvAkvar/AlphaTrade-4B-SFT-v0.1: LoRA-style
22 other repos hold byte-for-byte identical weight files.