SciEvalKit logo

SciEvalKit

InternScience/SciEvalKit

Unified evaluation toolkit and leaderboard for assessing scientific intelligence

GraphCanon updated Oct 18, 2026 · GitHub synced Oct 18, 2026

87 stars13 forksLast push Aug 30, 2026 Python Apache-2.0

Decision brief

Unverified catalog guidance: SciEvalKit is a unified evaluation toolkit and leaderboard designed to rigorously assess the scientific capabilities of large language and vision-language models throughout research processes.

Good fit when

  • Unverified catalog guidance: When assessing the scientific intelligence of multimodal models specifically across research stages
  • Unverified catalog guidance: If you need an integrated approach to evaluate models on various tasks within scientific inquiry

Avoid when

  • Unverified catalog guidance: For evaluating general performance without a focus on scientific applications and methodologies
  • Unverified catalog guidance: If your project does not benefit from an evaluation framework centered around vision-language abilities in scientific contexts

Observed Jul 16, 2026 · Source: enrich:decision_facts

Verify the decision

Maintenance and security

Full trust report
Maintenance
Steady (49d since push)
As of Oct 18, 2026
Provenance
Not a fork · Organization account
As of Oct 18, 2026
Security (OSV)
1 low; additional observations pending review
As of Jul 15, 2026

Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.

Install

Package installation is not established from the available documentation. Check the source repository for setup instructions.

git clone https://github.com/internscience/scievalkit
Source repository

Similar tools

Same-category neighbours. No typed graph edges are catalogued for this tool yet.

Evidence and technical details

Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.

Overview

SciEvalKit is a tool designed to rigorously evaluate the scientific capabilities of large language and vision-language models across every stage of the research process.

Capability facts

MCP server
Unknown; MCP server not established by this scan

Source: repo_scan · Oct 18, 2026

Languages
python

Source: github.language+scieval/dataset/PHYSICS/Math-Verify/pyproject.toml+scieval/dataset/PHYSICS/latex2sympy2_extended/pyproject.toml · Oct 18, 2026

Categories

Compatibility

Sourced claims from the README excerpt - not unsourced marketing copy.

Documented Python runtime (not runtime-tested)Python

Source: README excerpt (readme_declaration_v2, Oct 18, 2026)

pip install -e .[all] # brings in vllm, openai‑sdk, hf_hub, etc.
Source link

Tags

README

1 · Install

For agents

This page has a .md twin and JSON over the API.

Was this helpful?

Anonymous feedback helps us improve pages and translations.