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 SFT 2605
AI
US

UltraData SFT 2605

by OpenBMB

UltraData-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.

Dataset·Free

View on Hugging Face(opens in a new tab)

About UltraData SFT 2605

Key features of UltraData SFT 2605

  • Over 15 million SFT samples
  • Deep Thinking and Non-thinking variants for most domains
  • Seven domains including math, code, knowledge and instruction following
  • Multilingual math and knowledge data, Non-thinking only
  • Six-step quality pipeline
  • Benchmark decontamination of all data
  • Apache 2.0 license

Who UltraData SFT 2605 is for

  • Teams fine-tuning small language models
  • Researchers studying reasoning and non-reasoning SFT data
  • Developers building post-training data mixes

How to use UltraData SFT 2605

  1. 1

    Step 1: Sign in to Hugging Face and accept the access conditions on the dataset page.

  2. 2

    Step 2: Install the datasets library.

  3. 3

    Step 3: Load the config for the domain you want, such as math or code.

  4. 4

    Step 4: Choose the think or no_think data depending on whether you want reasoning traces.

  5. 5

    Step 5: Mix the subsets into your SFT run and hold out your own evaluation set.

UltraData SFT 2605 FAQ

What license is UltraData-SFT-2605 under?

Apache 2.0. You still need to accept the access conditions on Hugging Face.

What is the difference between think and no_think data?

Deep Thinking data targets reasoning, planning and verification on complex tasks. Non-thinking data targets fast, direct answers.

How is the data checked?

It goes through query filtering, answer quality filtering, validation training and decontamination against existing benchmarks.

Details

Pricing
FreeFree under Apache 2.0. Access requires a Hugging Face login and agreeing to share your Hugging Face information.
Type
Dataset
Categories
AIDataResearch
Added
Sep 17, 2026
OpPublisherOpenBMB

Built from public sources

We put this page together from UltraData SFT 2605’s own website and docs so you can compare it with other tools. Ploba isn’t affiliated with UltraData SFT 2605.

Work at UltraData SFT 2605? 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 · FreeUMUltraData MathOpenBMB · FreeUltraData-Math is an OpenBMB math pre-training dataset of over 290B tokens in three quality tiers, built to improve mathematical reasoning in language models.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
A2AgiBotWorld 2026AgiBot · FreeAgiBot World 2026 is a real-world robot dataset collected on the AGIBOT G2 platform in commercial and home settings. It is for researchers working on embodied intelligence and imitation learning.AI · FreeGAGAIAGAIA Benchmark · FreeGAIA is a benchmark of more than 450 questions with unambiguous answers for evaluating LLMs that use tools, search and prompting. It is for teams measuring how well AI assistants handle multi-step tasks.AI · FreeQoQokedasQokedas · Contact for pricingTraining-data company that turns real-world scientific signals and observations that are not normally written down into datasets for improving AI models in science.AI · Contact for pricing