What capabilities does DS-1000 test?

DS-1000 evaluates code generation.

How can a model improve its DS-1000 score?

Tools linked to DS-1000 on Sophon include OpenEnv Jupyter Agent (E2B-backed) - RL environments, datasets, and scaffolds that target this eval.

What license is DS-1000 under?

DS-1000 is available under CC-BY-SA-4.0.

DS-1000

Active

1,000 StackOverflow data-science questions covering NumPy/Pandas/Matplotlib/PyTorch/SciPy/SKLearn/TensorFlow - graded by hidden unit tests.

Open

Publisher: University of California, Berkeley
Capabilities: Code Generation
Domain: code
Format: HF Dataset
Size: 1000 tasks
License: CC-BY-SA-4.0
Published: May 2026
Notable for: Benchmark for evaluating code generation in the code domain.
Canonical: ds1000-code-gen.github.io
Also on: huggingface.co/datasets/xlangai/DS-1000

Cite

Notes

Only stored in your browser.

Related tools

View all

Implementations, trainers, datasets and scaffolds linked to this eval.

OpenEnv Jupyter Agent (E2B-backed)

Hugging Face

Stateful Jupyter-notebook environment backed by E2B sandboxes; the model edits and executes cells turn-by-turn, building up a notebook that solves a data-science task.

Trains towardRL EnvCode GenerationPlanningTool Calling

FAQ

What is DS-1000?: 1,000 StackOverflow data-science questions covering NumPy/Pandas/Matplotlib/PyTorch/SciPy/SKLearn/TensorFlow - graded by hidden unit tests.
What capabilities does DS-1000 test?: DS-1000 evaluates code generation.
How can a model improve its DS-1000 score?: Tools linked to DS-1000 on Sophon include OpenEnv Jupyter Agent (E2B-backed) - RL environments, datasets, and scaffolds that target this eval.
What license is DS-1000 under?: DS-1000 is available under CC-BY-SA-4.0.