Agentic AI Atlasby a5c.ai
/
GitHubDocsDiscord
iiRecord
Agentic AI Atlas · Inspect AI
eval-harness:inspect-aia5c.ai
II.
EvalHarness overview

eval-harness:inspect-ai

Reference · live

Inspect AI overview

UK AISI's evaluation framework for LLM agent and capability evals. Solver/scorer abstraction; first-class log and replay format.

EvalHarnessOutgoing · 0Incoming · 1

Attributes

displayName
Inspect AI
harnessKind
inspect-ai
homepageUrl
https://github.com/UKGovernmentBEIS/inspect_ai
description
UK AISI's evaluation framework for LLM agent and capability evals. Solver/scorer abstraction; first-class log and replay format.

Outgoing edges

None.

Incoming edges

uses_harness1