← all releases
openbmb·Feb 10, 2026·v1.0·open source

UltraData-Math

UltraData-Math is a large-scale, high-quality mathematical pre-training dataset designed to enhance mathematical reasoning in large language models (LLMs). It comprises over 290 billion tokens across three progressive tiers: L1 (170.5B tokens from web corpus), L2 (33.7B tokens from quality-selected data), and L3 (88B tokens from multi-format refined data). This dataset has been utilized in the mathematical pre-training of the MiniCPM Series models.

ML
visit website
overview

UltraData-Math is a large-scale, high-quality mathematical pre-training dataset designed to enhance mathematical reasoning in large language models (LLMs). It comprises over 290 billion tokens across three progressive tiers: L1 (170.5B tokens from web corpus), L2 (33.7B tokens from quality-selected data), and L3 (88B tokens from multi-format refined data). This dataset has been utilized in the mathematical pre-training of the MiniCPM Series models.

key features
  • 01Comprehensive dataset with over 290 billion tokens
  • 02Three progressive tiers (L1, L2, L3) for systematic enhancement
  • 03Applied in pre-training MiniCPM Series models
related productsbrowse all →
SkyRL txopen sourceSkyRL tx is an open-source library that implements a backend for the Tinker API, enabling users to set up their own Tinker-like services on personal hardware. It supports end-to-end reinforcement learning (RL) and offers significantly faster sampling. The library is designed to be modular, allowing easy prototyping of new training algorithms, environments, and execution plans without compromising usability or speed.Step-Audio-R1open sourceStep-Audio-R1 is an advanced audio language model developed by StepFun AI, designed to enhance audio reasoning capabilities by grounding its reasoning in acoustic features. It introduces Modality-Grounded Reasoning Distillation (MGRD), an iterative training framework that shifts the model's reasoning from textual abstractions to acoustic properties, effectively addressing the 'inverted scaling' problem where performance degrades with longer reasoning. This model has demonstrated superior performance across various audio understanding and reasoning benchmarks, surpassing models like Gemini 2.5 Pro and achieving results comparable to Gemini 3 Pro.SINQopen sourceSINQ (Sinkhorn-Normalized Quantization) is a novel, fast, and high-quality quantization method designed to make any Large Language Model (LLM) smaller while preserving accuracy. It offers a plug-and-play, model-agnostic technique that delivers state-of-the-art performance for LLMs without sacrificing accuracy.