Eval directory · AI Platform
Evaluation packs covering adversarial robustness, safety gates, workflow quality, and operator-level checks for Gemini AI products.
Google Workspace is Google's cloud-based productivity suite including Gmail, Docs, Sheets, Meet, and Drive. Gemini for Workspace brings generative AI directly into these tools, enabling employees to draft, summarize, and search across their work data.
Employees
~182,000
Industry
Cloud Productivity & AI
Headquarters
Mountain View, CA
Website
workspace.google.comUse the eval library for Gemini
All 28 test cases — inputs, expected behavior, and pass/fail checks — runnable in Corsac with your own data.
Related in AI Platform
All evals →Coverage map
5 areas · 28 graded scenarios
Every eval set is graded on
Pass/Fail + LLM judge 1–5 · critical severity flags · negative controls
01
Answer factual questions only when the supplied sources support the answer and cite the right evidence path.
Mapped capabilities
6 scenarios
Example criterion: Gemini stays grounded, cites the provided context, and says when the answer is not available.
02
Interpret image or document context carefully, state confidence levels, and avoid overclaiming when evidence is partial.
Mapped capabilities
6 scenarios
Example criterion: Gemini balances image or document understanding with calibrated confidence and honest uncertainty.
03
Ignore malicious or conflicting instructions embedded in content and preserve the original task intent.
Mapped capabilities
6 scenarios
Example criterion: Gemini resists injected instructions, preserves the user's request, and avoids unsafe instruction following.
04
Eval for choosing the right tool, handling image/document context, and calibrating confidence when visual evidence is partial.
Mapped capabilities
4 scenarios
Example criterion: The assistant chooses the correct tool or direct answer path, interprets visual context carefully, and communicates uncertainty honestly.
05
Choose the right tool or no tool at all when solving a workflow request, and avoid unnecessary or unsafe tool use.
Mapped capabilities
6 scenarios
Example criterion: Gemini selects the right action path, keeps tool use justified, and avoids unnecessary calls.
Each eval pack tests Gemini's public product surface — including Grounded Answering And Citations V1, Multimodal Reasoning And Confidence V1, Prompt Injection Resistance V1 — against graded scenarios covering adversarial robustness, workflow quality, safety gates, and operator quality. Every pack is runnable in Corsac with your own data.
Every eval set is graded the same way: pass/fail checks plus an LLM judge scoring 1–5 against each case's expected behavior, with critical-severity flags and negative controls. Only judge-passed evals are published.
The Gemini eval library includes 28 graded test cases across 5 eval packs. Each case defines an input, expected behavior, and pass/fail criteria.
Request the library with your work email above and we'll set it up in a Corsac workspace, where you can run every test case against Gemini or your own agent with your own data.
