datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
OpenThoughts-1k-sample
[!NOTE]
We have released a paper for OpenThoughts! See our paper here.
Open-Thoughts-1k-sample
This is a 1k sample of the OpenThoughts-114k dataset.
Open synthetic reasoning dataset with high-quality examples covering math, science, code, and puzzles!
Inspect the content with rich formatting with Curator Viewer.
Available Subsets
default subset containing ready-to-train data used to finetune the OpenThinker-7B and OpenThinker-32B models:
ds =… See the full description on the dataset page: https://huggingface.co/datasets/ryanmarten/OpenThoughts-1k-sample.OpenThoughts-114k
[!NOTE]
We have released a paper for OpenThoughts! See our paper here.
Open-Thoughts-114k
Open synthetic reasoning dataset with 114k high-quality examples covering math, science, code, and puzzles!
Inspect the content with rich formatting with Curator Viewer.
Available Subsets
default subset containing ready-to-train data used to finetune the OpenThinker-7B and OpenThinker-32B models:
ds = load_dataset("open-thoughts/OpenThoughts-114k"… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/OpenThoughts-114k.OpenThoughts3-1.2M
paper |
dataset |
model
[!NOTE]
We have released a paper for OpenThoughts! See our paper here.
OpenThoughts3-1.2M
Open-source state-of-the-art reasoning dataset with 1.2M rows. 🚀
OpenThoughts3-1.2M is the third iteration in our line of OpenThoughts datasets, building on our previous OpenThoughts-114k and OpenThoughts2-1M.
This time around, we scale even further and generate our dataset in a much more systematic way -- OpenThoughts3-1.2M is the result of a… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/OpenThoughts3-1.2M.Openthoughts_math_30k_opsdTaskTrove
TaskTrove
v5.1 (current) — independent-review source retirement — moves 15 sources with majority or unanimous REJECT verdicts out of the default config and into deprecated/. Three blinded reviewers each sampled 10 tasks per source from all 50 v5.0 source-drop candidates, read the instructions and packaged tests, and issued independent KEEP or REJECT verdicts. The 15 retired sources received at least two REJECT votes. The active catalog changes from 93 sources and 1,674,033… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/TaskTrove.AgentTrove
AgentTrove
AgentTrove is the largest open-source collection of agentic interaction traces to date, released by the OpenThoughts-Agent team. It contains 1,696,847 rows drawn from 219 source datasets spanning code repair, shell scripting, mathematical problem-solving, competitive programming, and general computer-use tasks.
At 1.7 million rows, AgentTrove is 4× the size of the Nemotron Terminal Corpus (430 K rows), the previous largest open-source agentic trace dataset.… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/AgentTrove.OpenThoughts2-1M
[!NOTE]
We have released a paper for OpenThoughts! See our paper here.
OpenThoughts2-1M
Open synthetic reasoning dataset with 1M high-quality examples covering math, science, code, and puzzles!
OpenThoughts2-1M builds upon our previous OpenThoughts-114k dataset, augmenting it with existing datasets like OpenR1, as well as additional math and code reasoning data.
This dataset was used to train OpenThinker2-7B and OpenThinker2-32B.
Inspect the content with rich… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/OpenThoughts2-1M.OpenThoughts-114k-mathThis is a filtered and metadata enriched version of open-thoughts/OpenThoughts-114k.
While the original dataset is a valuable resource containing DeepSeek-R1 outputs, it has very little metadata (only 2 fields: system and conversations). It does not contain, for instance, the original solution label, which means that we can not verify the model answers.
What we did
filtered the dataset for math content (math questions were prefixed by "Return your final response within… See the full description on the dataset page: https://huggingface.co/datasets/open-r1/OpenThoughts-114k-math.OpenThoughts-TB-dev
Project |
SFT dataset |
RL dataset |
SFT model
RL model
OpenThoughts-TB-Dev
Dataset Description
OpenThoughts-TB-Dev is our development benchmark for evaluating agent models on terminal and shell-based tasks. This dataset was curated to measure the effectiveness of different data sources, teacher models, and curation approaches during the development of OpenThinker-Agent-v1. We used this benchmark to ablate over many different instruction generation… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/OpenThoughts-TB-dev.OpenThoughts-TBLite
Blog Post |
GitHub |
Dev Set v1
OpenThoughts-TBLite
A Difficulty-Calibrated Benchmark for Building Terminal Agents
By OpenThoughts Agent team, Snorkel AI, Bespoke Labs
OpenThoughts-TBLite is a curated collection of 100 Terminal-Bench tasks that closely track TB2 performance, but run much faster. It's designed to be more informative during model development, making it ideal for debugging, iteration, and training ablations.
Dataset Description… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/OpenThoughts-TBLite.OpenThoughts-Agent-v1-SFT
Project |
SFT dataset |
RL dataset |
SFT model |
RL model
OpenThinker-Agent-v1-SFT
OpenThoughts-Agent is an open-source effort to curate the best datasets for training agents. Our first release includes datasets, models and our research codebase.
OpenThinker-Agent-v1 is a model trained for agentic tasks such as Terminal-Bench 2.0 and SWE-Bench.
The OpenThinker-Agent-v1 model is post-trained from Qwen/Qwen3-8B.
It is SFT-ed on the OpenThoughts-Agent-v1-SFT dataset, then… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/OpenThoughts-Agent-v1-SFT.OpenThoughts-TB-dev-v2
Blog Post |
GitHub |
Dev Set v1
OpenThoughts-TB-dev-v2
A Difficulty-Calibrated Benchmark for Building Terminal Agents
By OpenThoughts Agent team, Snorkel AI, Bespoke Labs
OpenThoughts-TB-dev-v2 is a curated collection of 100 Terminal-Bench tasks that closely track TB2 performance, but run much faster. It's designed to be more informative during model development, making it ideal for debugging, iteration, and training ablations.
Dataset Description… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/OpenThoughts-TB-dev-v2.openthoughts_18K_solutions_R1_distill_Llama_8BOpenThoughts-1k-SampledOpenThoughts-Agent-SFT-100K
Project |
Code |
Collection
OpenThoughts-Agent-SFT-100K
OpenThoughts-Agent is an open-source effort to curate the best datasets for training agents. Our release includes datasets, models and our research codebase.
OpenThoughts-Agent-SFT-100K is the 100,000-example point of the OpenThoughts-Agent SFT scaling ladder (sizes 316 / 1K / 3.16K / 10K / 31.6K / 100K). It contains (task, agent-trajectory) pairs used to fine-tune OpenThinkerAgent-8B-SFT-100K and… See the full description on the dataset page: https://huggingface.co/datasets/open-thoughts/OpenThoughts-Agent-SFT-100K.openthoughts-4-math-qwen3-32b-7k-annotated-sharegptopenthoughts_science_1kOpenThoughts-1k-Mapped-Sampledopen-thoughts-scienceOpenThoughts-TBLite
Blog Post |
GitHub |
Dev Set v1
OpenThoughts-TBLite
A Difficulty-Calibrated Benchmark for Building Terminal Agents
By OpenThoughts Agent team, Snorkel AI, Bespoke Labs
OpenThoughts-TBLite is a curated collection of 100 Terminal-Bench tasks that closely track TB2 performance, but run much faster. It's designed to be more informative during model development, making it ideal for debugging, iteration, and training ablations.
Dataset Description… See the full description on the dataset page: https://huggingface.co/datasets/shanxianzheng/OpenThoughts-TBLite.openthoughts3-en-ar-midtrain
openthoughts3-en-ar-midtrain
Arabic translation of the OpenThoughts3_1.2M split of smoltalk2 (config Mid): long mathematical reasoning traces with <think> blocks, in a two-message user/assistant format. Translated with google/gemma-4-12B-it (bf16, greedy) on A100s. All 1,135,104 source rows are present, none dropped.
The pipeline segments each message into prose and verbatim blocks (code, LaTeX, tables, and inline non-translatables are masked and never sent to the model)… See the full description on the dataset page: https://huggingface.co/datasets/SultanR/openthoughts3-en-ar-midtrain.open-thoughts-4-math-qwen3-32b-annotated
Dataset Card for Open-Thoughts-4-Math-Qwen3-32B-Annotated
This dataset is the Qwen3-32B annotated version of mlfoundations-dev/hero_run_4_math curated by the
OpenThoughts4 team. We provide the responses from Qwen3-32B in the generated_text column. These samples were generated using temperature = 0.8 and max output tokens = 7,500.
We note that many of the responses are truncated, so use this dataset wisely!
Dataset Details
Dataset Description… See the full description on the dataset page: https://huggingface.co/datasets/marin-community/open-thoughts-4-math-qwen3-32b-annotated.OpenThoughts-TB-DevOpenThoughts-114k-math-correct-qwen3-14b-math-prepared-topk128-normalizedOpenThoughts3edition_0558_ryanmarten-OpenThoughts-1k-sample-readymade
edition_0558_ryanmarten-OpenThoughts-1k-sample-readymade
A Readymade by TheFactoryX
Original Dataset
ryanmarten/OpenThoughts-1k-sample
Process
This dataset is a "readymade" - inspired by Marcel Duchamp's concept of taking everyday objects and recontextualizing them as art.
What we did:
Selected the original dataset from Hugging Face
Shuffled each column independently
Destroyed all row-wise relationships
Preserved structure, removed meaning
The result:
Same… See the full description on the dataset page: https://huggingface.co/datasets/TheFactoryX/edition_0558_ryanmarten-OpenThoughts-1k-sample-readymade.OpenThoughts-114k-Code_decontaminated
Dataset description
This dataset is the same as open-r1/OpenThoughts-114k-Code decontaminated against the benchmark datasets.
The decontamination has been run using the script in huggingface/open-r1:
python scripts/decontaminate.py \
--dataset "open-r1/OpenThoughts-114k-Code" \
-c
...
Removed 2 samples from 'aime_2025'
Removed 28 samples from 'math_500'
Removed 3482 samples from 'lcb'
Initial size: 19890, Final size: 16378
openthoughts3_code_100k_annotated_QwQ-32B_sharegpt_eval_5554
mlfoundations-dev/openthoughts3_code_100k_annotated_QwQ-32B_sharegpt_eval_5554
Precomputed model outputs for evaluation.
Evaluation Results
Summary
Metric
AIME24
AMC23
MATH500
MMLUPro
JEEBench
GPQADiamond
LiveCodeBench
CodeElo
CodeForces
HLE
HMMT
AIME25
LiveCodeBenchv5
Accuracy
34.3
74.5
79.4
49.4
51.0
44.3
53.9
21.5
23.1
12.2
17.0
22.7
40.1
AIME24
Average Accuracy: 34.33% ± 1.89%
Number of Runs: 10
Run
Accuracy
Questions… See the full description on the dataset page: https://huggingface.co/datasets/mlfoundations-dev/openthoughts3_code_100k_annotated_QwQ-32B_sharegpt_eval_5554.open_thoughts3_1.2m_annotated_Qwen3-8B_sft_packingopenthoughts-4-code-qwen3-32b-32k-annotated