Qwen/Qwen2.5-VL-3B-Instruct
216 derivatives checked
- Qwen/Qwen2.5-VL-3B-Instruct-AWQ: Quantization
- nanonets/Nanonets-OCR2-3B: LoRA-style
- ggml-org/Qwen2.5-VL-3B-Instruct-GGUF: Same weights, converted (fewer layers)
- typhoon-ai/typhoon-ocr-3b: Full fine-tune
- nanonets/Nanonets-OCR-s: LoRA-style
- stanford-oval/churro-3B: Full fine-tune
- huihui-ai/Qwen2.5-VL-3B-Instruct-abliterated: Abliterated
- RedHatAI/Qwen2.5-VL-3B-Instruct-quantized.w8a8: Quantization
- unsloth/Qwen2.5-VL-3B-Instruct-GGUF: Same weights, converted (fewer layers)
- lmstudio-community/Qwen2.5-VL-3B-Instruct-GGUF: Quantization (fewer layers)
- tsystems/colqwen2.5-3b-multilingual-v1.0: Adapter
- sherif1313/Arabic-handwritten-OCR-4bit-Qwen2.5-VL-3B-v2: Quantization
- IDEA-Research/Rex-Omni: Full fine-tune
- nomic-ai/nomic-embed-multimodal-3b: Adapter
- mradermacher/Qwen2.5-VL-3B-Instruct-GGUF: Same weights, converted (fewer layers)
- RedHatAI/Qwen2.5-VL-3B-Instruct-FP8-dynamic: Quantization
- tsystems/colqwen2.5-3b-base: Same weights, repacked
- unsloth/Qwen2.5-VL-3B-Instruct-bnb-4bit: Quantization
- unsloth/Qwen2.5-VL-3B-Instruct-unsloth-bnb-4bit: Quantization
- mlx-community/Qwen2.5-VL-3B-Instruct-4bit: Quantization
- unsloth/Qwen2.5-VL-3B-Instruct: Same weights, repacked
- numind/NuExtract-2.0-4B: Full fine-tune
- kyutai/CASA-Qwen2_5-VL-3B-LiveCC: Full fine-tune
- Mungert/Nanonets-OCR-s-GGUF: LoRA-style (fewer layers)
- Efficient-Large-Model/Fast_dVLM_3B: Partial fine-tune
- sherif1313/Arabic-English-handwritten-OCR-v3: LoRA-style
- Mungert/Holo1-3B-GGUF: Full fine-tune (fewer layers)
- Mungert/Nanonets-OCR2-3B-GGUF: LoRA-style (fewer layers)
- zss01/PixelCraft-3B: Full fine-tune
- DevQuasar-6/Qwen.Qwen2.5-VL-3B-Instruct-GGUF: Same weights, converted (fewer layers)
30 other repos hold byte-for-byte identical weight files.