InternScience/SciEvalKit

A unified evaluation toolkit and leaderboard for rigorously assessing the scientific intelligence of large language and vision–language models across the full research workflow.

46
/ 100
Emerging

This toolkit helps AI researchers and developers accurately measure how well large language and vision-language models perform on complex scientific tasks, not just general conversations. It takes a model and a set of scientific challenges (like interpreting images, symbolic reasoning, or generating code) and outputs a detailed score, revealing how scientifically intelligent the model truly is across different research workflow stages. Scientists, engineers, and AI developers building or using these advanced models would find this essential for rigorous evaluation.

Use this if you need to rigorously evaluate the scientific intelligence of large language or vision-language models across the entire research workflow, rather than relying on general-purpose benchmarks.

Not ideal if you are looking for a simple, quick way to test a model's basic conversational or broad-domain reasoning abilities.

AI-model-evaluation scientific-AI research-workflow-automation multimodal-AI scientific-computing
No Package No Dependents
Maintenance 10 / 25
Adoption 9 / 25
Maturity 13 / 25
Community 14 / 25

How are scores calculated?

Stars

74

Forks

10

Language

Python

License

Apache-2.0

Last pushed

Feb 27, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/InternScience/SciEvalKit"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.