allenai/Olmo-3-7B-Instruct
Full fine-tune vs allenai/Olmo-3-7B-Instruct-DPO. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, MLP 96/96, Norms 128/129, Attention 128/128, lm_head 1/1
34 derivatives checked
- allenai/Molmo2-O-7B: Different architecture
- unsloth/Olmo-3-7B-Instruct-GGUF: Same weights, converted
- bartowski/allenai_Olmo-3-7B-Instruct-GGUF: Same weights, converted
- prithivMLmods/Olmo-3-7B-Instruct-AIO-GGUF: Same weights, converted
- lmstudio-community/Olmo-3-7B-Instruct-GGUF: Quantization
- ChiKoi7/Olmo-3-7B-Instruct-Heretic-GGUF: Abliterated
- mradermacher/Olmo-3-7B-Instruct-i1-GGUF: Quantization
- Ex0bit/Elbaz-Olmo-3-7B-Instruct-abliterated: Abliterated
- unsloth/Olmo-3-7B-Instruct-unsloth-bnb-4bit: Quantization
- eaddario/Olmo-3-7B-Instruct-GGUF: Same weights, converted
- mradermacher/Olmo-3-7B-Instruct-GGUF: Same weights, converted
- mlx-community/Olmo-3-7B-Instruct-4bit: Quantization
- mlx-community/Olmo-3-7B-Instruct-mxfp4-QAT: Quantization
- unsloth/Olmo-3-7B-Instruct: Identical copy
- Tuwhy/Olmo-3-7B-Instruct-OPSA-Code: Full fine-tune
- jkim96/Olmo-3-7B-Instruct-DASHQ-Q2-GGUF: Quantization
- mlx-community/Olmo-3-7B-Instruct-8bit: Quantization
- noctrex/Olmo-3-7B-Instruct-abliterated-GGUF: Abliterated
- h3rb3rn/moe-sovereign-planner-olmo3-7b: Quantization
- jimfhahn/bibframe-olmo3-v3: LoRA-style
- MaziyarPanahi/Olmo-3-7B-Instruct-GGUF: Same weights, converted
- CCATresearch/Olmo-3-7B-Instruct: Same weights, converted
- Tuwhy/Olmo-3-7B-Instruct-OPSA: Full fine-tune
- Harvard-DCML/ADAPT-Olmo3-4.3B-Instruct: Full fine-tune (fewer layers)
- vanta-research/atom-olmo3-7b: LoRA-style
- Goekdeniz-Guelmez/Josiefied-Olmo-3-7B-Instruct-abliterated-v1: LoRA-style
- xzuyn/Olmo-3-7B-Instruct-TrashMix-v1: LoRA-style
- ChiKoi7/Olmo-3-7B-Instruct-Heretic: Abliterated
- ConicCat/Role-mo-V2-7B: LoRA-style
- ConicCat/Role-mo-V3-7B: LoRA-style
2 other repos hold byte-for-byte identical weight files.