OLMo

OLMo provides downloadable, fully open-source transformer model weights in 7B and 32B parameter variants with complete architectural transparency. Users can deploy these models locally or via APIs without proprietary restrictions, with all training code, data, and evaluation artifacts publicly available for reproducibility and modification. The model family includes base, instruction-tuned, and reasoning-focused variants enabling different use cases from raw text generation to multi-turn dialogue.

OLMo-32B-Instruct and 7B-Instruct variants are post-trained using supervised fine-tuning (SFT) and direct preference optimization (DPO) on instruction-following and dialogue corpora. These models support multi-turn conversation context, tool calling for function invocation, and structured response generation. The instruction tuning pipeline is fully documented and reproducible via the Open Instruct framework, allowing users to understand and modify training data composition.

OLMo provides direct download of model weights in standard formats, enabling users to deploy models locally without cloud dependencies or API keys. Model weights are available for all variants (7B, 32B, base, instruct, think) and can be used with standard inference frameworks. This approach provides maximum control, privacy, and reproducibility for deployment.

OLMo-32B-Think and 7B-Think variants are trained to generate intermediate reasoning steps before producing final answers, using supervised fine-tuning (SFT), direct preference optimization (DPO), and reinforcement learning (RL) on reasoning-focused data. These models decompose complex problems into step-by-step reasoning traces, enabling better performance on math, logic, and multi-step reasoning tasks. The thinking training pipeline is fully reproducible via Open Instruct.

OLMo provides OlmoCore, a fully open training framework enabling users to reproduce the original training runs or fine-tune models on custom data. The framework supports configuration-driven training with documented hyperparameters, data mixing strategies, and training stages (pretraining, mid-training, instruction tuning, DPO, RL). Users can access training code, training data artifacts, and training logs for complete reproducibility and modification.

OLMo provides Duplodocus, a fuzzy deduplication tool, and Datamap-rs, a large-scale data cleaning utility, as open-source components used in the training pipeline. These tools enable users to preprocess training data at scale, removing duplicates and low-quality examples before training. The tools are designed for web-scale datasets and are fully reproducible, allowing researchers to understand and audit data quality decisions.

OLMo provides OlmoTrace, a tool for attributing model outputs and behaviors to specific training examples or data sources. This enables users to trace which training documents influenced particular model predictions, supporting interpretability research and data auditing. The tool works by analyzing model attention patterns and gradient information to identify influential training examples, providing transparency into model decision-making.

OLMo provides OLMES, a reproducible evaluation utility for assessing model performance on standardized benchmarks. OLMES enables users to evaluate OLMo models (or other models) on consistent, documented evaluation protocols, supporting research reproducibility and fair model comparison. The evaluation framework is fully open-source and includes benchmark datasets, evaluation scripts, and metric computation.

OLMo provides Decon, a tool for detecting and removing test set contamination from training data. This tool identifies training examples that overlap with evaluation benchmarks, preventing inflated performance metrics and ensuring fair model evaluation. Decon enables users to audit training data for benchmark contamination and remove problematic examples before training.

OLMo provides FlexOlmo, a collaborative training paradigm enabling distributed training across multiple organizations or compute providers. FlexOlmo allows participants to contribute compute resources and data to jointly train models, with transparent accounting of contributions and fair reward distribution. This approach enables resource-constrained teams to participate in large-scale model training.

OLMo provides a web-based chat interface ('Chat with Olmo') enabling users to interact with OLMo models through a browser without local setup or API keys. The interface supports multi-turn conversation, streaming responses, and real-time interaction. This provides an accessible entry point for non-technical users and researchers to explore model capabilities.