nvidia/OpenScienceReasoning-2 download history

nvidia/OpenScienceReasoning-2 is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 17,549 times (3,805 in the last 7 days), and 50,756 times in total. It ranks #1,780 among datasets by monthly downloads.

Dataset Description OpenScienceReasoning-2 is a multi-domain synthetic dataset designed to improve general-purpose reasoning in large language models (LLMs). The dataset contains multiple-choice and open-ended question-answer pairs with detailed reasoning traces and spans across diverse s

Models trained on OpenScienceReasoning-2

26 models list it as training data.

• mradermacher/Qwen3-4B-MegaR3ASONER-v1-i1-GGUF 1.1K downloads in 30 days
• mradermacher/OpenScienceReasoning-Qwen-e10-i1-GGUF 847 downloads in 30 days
• mradermacher/Qwen3-4B-MegaR3ASONER-v1-GGUF 601 downloads in 30 days
• mradermacher/SciReason-LFM2-2.6B-i1-GGUF 594 downloads in 30 days
• mradermacher/vanta-research-apollo-v1-4b-GGUF 407 downloads in 30 days
• mratsim/MiniMax-M2.5-FP8-INT4-AWQ 388 downloads in 30 days
• mradermacher/OpenScienceReasoning-Qwen-e10-GGUF 307 downloads in 30 days
• mradermacher/SciReason-LFM2-2.6B-GGUF 201 downloads in 30 days
• mratsim/MiniMax-M2.5-BF16-INT4-AWQ 155 downloads in 30 days
• mratsim/Behemoth-X-123B-v2-NVFP4 143 downloads in 30 days
• mratsim/MiniMax-M2.1-FP8-INT4-AWQ 99 downloads in 30 days
• aisingapore/Nemotron-SEA-LION-v4.8-30B-A3B-Base 43 downloads in 30 days

Spaces using OpenScienceReasoning-2

• Model Pulse 8 likes

Open nvidia/OpenScienceReasoning-2 on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.