Ploba logo

Find the right AI tool for the job: apps, agents, APIs, MCP servers and datasets, explained plainly.

Get the AI Digest

Discover

  • Explore
  • Top Charts
  • Categories
  • Agents
  • MCP servers
  • Developer tools

Learn & community

  • Learn AI
  • AI Glossary
  • Blog
  • Hackathons
  • People
  • AI Digest

For makers

  • List your tool
  • Docs
  • Claim a listing
  • Advertise
  • Write for us

Company

  • About
  • Careers
  • Changelog
  • Trust
  • Security
  • Contact

© 2026 Ploba. All rights reserved.

  • Privacy
  • Terms
  • Cookies
  • Advertising
  • All policies
llms.txt
Ploba logo
DiscoverHackathonsPeopleLearnAI Digest
Sign inGet started
HomeTop ChartsExploreCategories
TypesWeb apps1,894Agents585Plugins416APIs376Mobile apps152Datasets85
AdvertisingYour product hereReach people browsing AI tools on Ploba.Advertise hereAdvertisingYour product hereReach people browsing AI tools on Ploba.Advertise here
  1. Explore
  2. /
  3. AI
  4. /
  5. UltraData Math
AI
UM

UltraData Math

by OpenBMB

UltraData-Math is an OpenBMB math pre-training dataset of over 290B tokens in three quality tiers, built to improve mathematical reasoning in language models.

Dataset·Free

View on Hugging Face(opens in a new tab)

About UltraData Math

Key features of UltraData Math

  • Over 290B tokens across three tiers
  • L1 web corpus of 170.5B tokens
  • L2 quality-selected subset of 33.7B tokens
  • L3 refined data in Q&A, dialogue and textbook formats
  • Math formulas standardized into LaTeX
  • Document-level deduplication
  • Mainly English and Chinese content

Who UltraData Math is for

  • Teams pre-training language models on math
  • Researchers studying data quality grading
  • Engineers building math reasoning models

How to use UltraData Math

  1. 1

    Step 1: Open the dataset page on Hugging Face and review the tier table.

  2. 2

    Step 2: Pick a tier: L1 for scale, L2 for quality-selected text, L3 for structured reasoning formats.

  3. 3

    Step 3: Load the chosen tier with the Hugging Face datasets library.

  4. 4

    Step 4: Mix it into your pre-training data and compare results on math benchmarks.

  5. 5

    Step 5: Read the linked paper for the tiered data management framework.

UltraData Math FAQ

What are the L1, L2 and L3 tiers?

L1 is a filtered web math corpus, L2 is a quality-selected subset of it, and L3 is rewritten or synthetic data in formats like Q&A, dialogue and textbooks.

How large is UltraData-Math?

More than 290B tokens in total: 170.5B in L1, 33.7B in L2 and 88B in L3.

What languages does it cover?

The card says the filtering aims for mainly English and Chinese mathematical content.

Has it been used to train models?

Yes. The card says it has been applied to mathematical pre-training of the MiniCPM series.

Details

Pricing
Free
Type
Dataset
Categories
AIDataEducation
Added
Sep 17, 2026
OpPublisherOpenBMB

Built from public sources

We put this page together from UltraData Math’s own website and docs so you can compare it with other tools. Ploba isn’t affiliated with UltraData Math.

Work at UltraData Math? Claim the listing to manage it yourself.

Something wrong here?

More from OpenBMB

View publisher
UlUltra-FineWebOpenBMB · FreeUltra-FineWeb is OpenBMB's filtered web pre-training dataset, with about 1T English tokens and 120B Chinese tokens. It is built by applying a verification-based quality filter to FineWeb and Chinese FineWeb data.AI · FreeUSUltraData SFT 2605OpenBMB · FreeUltraData-SFT-2605 is OpenBMB's set of over 15 million supervised fine-tuning samples used to post-train a small on-device language model. It covers math, code, knowledge and instruction following, with both deep-thinking and non-thinking data.AI · FreeUSUltraData SFT Agent 2609OpenBMB · FreeUltraData-SFT-Agent-2609 is OpenBMB's dataset of 483,661 agent trajectories for instruction tuning, covering tool use, search, code and general agent tasks. It keeps full trajectories with tool calls, observations and errors.Developer Tools · Free

Related tools

See more
MRMath ReasoningIFM · FreeMath-Reasoning from IFM is a dataset of math problem-solving, rewriting and dialogue data for training reasoning-oriented language models. It is one repository in the K2 Horizon dataset series.AI · FreeFineWeb-Edu iconFineWeb-EduHugging Face · FreeFineWeb-Edu is a 1.3 trillion token dataset of educational web pages filtered from Hugging Face's FineWeb using an educational-quality classifier. It is for teams pretraining or researching large language models.AI · FreeNemotron Math Proofs v3 SFT iconNemotron Math Proofs v3 SFTNVIDIA · FreeNemotron Math Proofs v3 SFT is an NVIDIA dataset of long-form proof generation, refinement, verification and meta-verification traces. It is for training language models on mathematical reasoning.AI · Free