allenai/olmOCR-2-7B-1025
Full fine-tune vs Qwen/Qwen2.5-VL-7B-Instruct. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 2/2, Norms 122/122, MLP 280/280, Attention 324/324, lm_head 1/1
13 derivatives checked
- lmstudio-community/olmOCR-2-7B-1025-GGUF: Quantization (fewer layers)
- bartowski/allenai_olmOCR-2-7B-1025-GGUF: Same weights, converted (fewer layers)
- mradermacher/olmOCR-2-7B-1025-GGUF: Same weights, converted (fewer layers)
- mradermacher/olmOCR-2-7B-1025-i1-GGUF: Quantization (fewer layers)
- richardyoung/olmOCR-2-7B-1025-GGUF: Quantization (fewer layers)
- Charitarth/olmOCR-2-7B-1025-GPTQ-W4A16: Quantization
- winninghealth/olmOCR-2-7B-1025-INT4: Quantization
- orgessnk/olmOCR-2-7B-1025-BNB-4bit: Quantization
- richardyoung/olmOCR-2-7B-1025-MLX-8bit: Quantization
- cstr/olmOCR-2-7B-1025-GGUF: Not comparable yet
- suv11235/olmOCR-7B-grpo-v2: Partial fine-tune
- suv11235/olmOCR-7B-grpo-v6: Partial fine-tune
- suv11235/olmOCR-7B-grpo-v3: Partial fine-tune
1 other repos hold byte-for-byte identical weight files.