Qwen/Qwen3-Reranker-4B
Full fine-tune vs Qwen/Qwen3-4B-Base. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Norms 145/145, MLP 108/108, Attention 144/144
29 derivatives checked
- boboliu/Qwen3-Reranker-4B-W4A16-G128: Quantization
- dolfsai/Qwen3-Reranker-4B-seq-cls-vllm-W4A16: Quantization
- Voodisss/Qwen3-Reranker-4B-GGUF-llama_cpp: Same weights, converted
- mradermacher/Qwen3-Reranker-4B-GGUF: Same weights, converted
- furiosa-ai/Qwen3-Reranker-4B: Identical copy
- danielchalef/Qwen3-Reranker-4B-seq-cls-vllm-fixed: Same weights, converted
- DCC-BS/Qwen3-Reranker-4B-FP8-Dynamic: Quantization
- giladgd/Qwen3-Reranker-4B-GGUF: Same weights, converted
- mradermacher/Qwen3-Reranker-4B-i1-GGUF: Quantization
- drawais/Qwen3-Reranker-4B-AWQ-INT4: Quantization
- DevQuasar/Qwen.Qwen3-Reranker-4B-GGUF: Same weights, converted
- vserifsaglam/Qwen3-Reranker-4B-4bit-MLX: Quantization
- IAAR-Shanghai/MemReranker-4B: Full fine-tune
- hq-bench/coreb-code-reranker: LoRA-style
- prithivMLmods/Qwen3-Reranker-4B-F32-GGUF: Same weights, converted
- dolfsai/Qwen3-Reranker-4B-seq-cls-vllm-W4A16_ASYM: Quantization
- tensorblock/Qwen_Qwen3-Reranker-4B-GGUF: Quantization
- KoliaNik/Qwen3-Reranker-4B-NVFP4: Quantization
- ebasKing/Qwen3-Reranker-4B-GGUF-llama_cpp: Same weights, converted
- tatsuyaaaaaaa/Qwen3-Reranker-4B-gguf: Same weights, converted
- gscoppino/Qwen3-Reranker-4B-GGUF-llama_cpp: Same weights, converted
- batiai/Qwen3-Reranker-4B-GGUF: Quantization
- MaYiding/EventTwin: LoRA-style
- Hanno-Labs/bosun-4b-GGUF: LoRA-style
- zenlm/zen-reranker-4B-GGUF: Quantization
- Lipdog/Qwen3-Reranker-4B-mlx-fp16: Same weights, converted
- soichisumi/qwen3-reranker-4b-mlx-4bit: Quantization
- zenlm/zen-reranker: Identical copy
- donghongjiang/SkillReason-reranker-4b: Full fine-tune
9 other repos hold byte-for-byte identical weight files.