StackMathQA Mathematics Question-Answer Corpus
Source, license and coverage
Supplier documentation. These claims are separate from the automated sample score. A listing edit date is not a data freshness date.
- License
- cc-by-4.0
- Source / creator
- math-ai/StackMathQA
- Collection method
- The authors sourced raw Q&A pairs from Stack Exchange's public data dumps across math-related sites, then filtered and curated for quality, deduplicated, and sharded into size-stratified configs (100k → 1.6M) plus a full 2M release. Math content retains its original LaTeX formatting. Per-question multiple answers are preserved as a list rather than collapsed.
- Coverage start
- Not documented
- Coverage end
- Not documented
- Data last updated
- Not documented
- Update schedule
- Not documented
Content is English-only and reflects the demographics and topic distribution of Stack Exchange users (skewed toward undergraduate/graduate pure math, with strong representation of algebra, calculus, and analysis; weaker on applied/numerical topics). Answer quality varies — high-vote answers are not separated from low-vote ones in all configs. LaTeX rendering assumes downstream MathJax/KaTeX support. The dataset is not deduplicated against common math benchmarks (MATH, GSM8K, MathQA), so benchmark contamination is a real risk for evaluation use. Source does not document a formal quality-filtering threshold; buyers should validate empirically.
Sample structure score: 100 / 100
This automated check describes the inspected sample, not factual accuracy, legal rights, representativeness, or the quality of the entire dataset. It is not a customer rating.
Assessed 10 sample records (JSON) on 2026-10-09. All records in the provided sample were checked.
| Check | Points | Evidence |
|---|---|---|
| Populated cells | 50 / 50 | 30 of 30 top-level cells contain a value. Null, absent and blank values count as missing; zero and false count as populated. |
| Consistent value types | 30 / 30 | 30 of 30 populated cells match their column's most common observed type. Types are inferred, not checked against real-world truth. |
| Consistent record shape | 20 / 20 | 10 of 10 records have the expected fields. CSV/TSV use the header width; JSON uses the union of observed keys. |
Field-level findings and improvements
Check missing cells and mixed types below. Document intentional missing values or mixed types in your field descriptions. Do not fill legitimate unknowns with invented values just to increase this score.
| Field | Missing cells | Most common type | Other populated types |
|---|---|---|---|
| Q | 0 / 10 | string | 0 / 10 |
| A | 0 / 10 | string | 0 / 10 |
| meta | 0 / 10 | object | 0 / 10 |
About this data
Mathematical questions and answers from Stack Exchange sites including Math, MathOverflow, and Stats. Suitable for math reasoning fine-tuning, pretraining, and retrieval-augmented generation tasks.
Retrieve with your agent or Python
Create an account and configure DATABAZAAR_API_KEY. This example retrieves free or already purchased data; it never makes a purchase. For a multi-file dataset, choose a file index from the manifest.
Download the Python examplepython3 retrieve-dataset.py 9b99632f-dff5-4939-91be-c5c86ef6ce96 --output dataset.bin
Full supplier documentation
Schema
| Name | Type | Description |
|---|---|---|
| Q | VARCHAR | Mathematical question text from Stack Exchange, potentially containing LaTeX-formatted equations and multi-line content. |
| A | VARCHAR | One or more answer texts from Stack Exchange responders, each potentially containing LaTeX-formatted mathematical content. |
| meta | STRUCT("language" VARCHAR, url VARCHAR, "timestamp" TIMESTAMP, "source" VARCHAR, question_score VARCHAR, answer_count BIGINT, answer_id BIGINT) | Metadata object containing source site, post URLs, timestamps, question/answer IDs, vote scores, and answer counts from Stack Exchange. |
Sample Data
Preview a sample of the data before downloading.
Public sample only. Sign in to retrieve the full dataset, including free datasets.
For AI Agents
# 1. Add to your agent's MCP config (claude_desktop_config.json or similar):
{
"mcpServers": {
"databazaar": { "command": "npx", "args": ["databazaar-mcp"] }
}
}
# 2. Your agent can then call:
search_datasets({ query: "StackMathQA Mathematics Questi" })
// Found: 9b99632f-dff5-4939-91be-c5c86ef6ce96
get_download_url({ dataset_id: "9b99632f-dff5-4939-91be-c5c86ef6ce96" }) // free — sign in with MCP OAuth first# Free dataset — sign in or use your account API key: curl https://api.databazaar.io/datasets/9b99632f-dff5-4939-91be-c5c86ef6ce96/download-url -H "Authorization: Bearer $DATABAZAAR_API_KEY"