Dolci
Allen Institute for AIAi2's post-training data suite for OLMo 3, covering supervised fine-tuning, DPO preference and RLVR sets for both the Instruct and Think variants. It is the successor to the Tulu mixtures, folding in OpenThoughts3, FLAN v2, OpenAssistant, WildChat and new Ai2 synthetic data.
A family record covering the Dolci-Instruct and Dolci-Think SFT, DPO and RL sets. Verified 2026-08-13 via the allenai/Dolci-Instruct-SFT dataset card on Hugging Face.
Openness
5 high confidence- license
- ODC-BY
- gated
- false
- dataset_card
- present
ODC-BY, ungated, per-stage dataset cards.
- https://huggingface.co/datasets/allenai/Dolci-Instruct-SFT recorded 2026-08-13
ODC-BY license, ungated, dataset card
Adoption
2 high confidence3,807 downloads in the trailing 30 days for allenai/Dolci-Instruct-SFT, which bands at 1K-10K, level 2 on the dataset adoption scale.
- https://huggingface.co/api/datasets/allenai/Dolci-Instruct-SFT recorded 2026-08-13
3,807 downloads in the trailing 30 days for allenai/Dolci-Instruct-SFT
Capability
5 high confidenceThe current fully-open frontier post-training suite; OLMo 3-Think 32B leads the fully-open thinking-model class in documented benchmarks. The Dolci card presents the suite as the successor to the Tulu mixtures.
- https://allenai.org/blog/olmo3 recorded 2026-08-13
OLMo 3-Think 32B (trained with Dolci) is the strongest fully-open thinking model: MATH 96.1, HumanEval+ 91.4, IFEval 89.0
- https://huggingface.co/datasets/allenai/Dolci-Instruct-SFT recorded 2026-08-13
Dolci is Ai2's OLMo 3 post-training suite, successor to the Tulu mixtures
Verified 2026-08-13