index / confident-ai/deepeval
deepeval
The LLM Evaluation Framework
IntegrationsFeature ScoringListedPythonreadme pick
Repository
Stars18,901
LanguagePython
LicenseApache-2.0
Created2023-08-10
Last pushed2026-10-15
Officialno
Judgment
Genuine Jev project0.45
gate 0.5
Uses Jev at runtime0.45
Is a meta list0.04
Is a reimplementation0.12
Quality scales
Substance2.99 / 3
gate 0.5
Docs2.01 / 3
Novelty1.73 / 3
Composite score0.93
Category distribution
Research and Evals0.68
Other0.21
Integrations0.07
SDKs and Clients0.02
Agent and Dev Tooling0.01
Applications0.01
Games and Simulation0.00
Learning0.00
Assigned categoryIntegrations
Category probability0.68
Category confidence0.63
PatternFeature Scoring
Pattern confidence0.95
Curation
StatusListed
Reasonoverride: Ships a first-class TypeSafe model adapter (deepeval/models/system_one/typesafe_model.py, TypeScript port in typescript/src/models/system-one, docs/content/integrations/models/typesafe-ai.mdx) so any metric can use Jev as its judge model
README pickyes
Override{"status":"listed","category":"integration","reason":"Ships a first-class TypeSafe model adapter (deepeval/models/system_one/typesafe_model.py, TypeScript port in typescript/src/models/system-one, docs/content/integrations/models/typesafe-ai.mdx) so any metric can use Jev as its judge model","by":"rhc"}
Sources
- code:"typesafe_sdk"
- code:"jev-latest"
- code:"@typesafe-ai/sdk"
Provenance
Modeljev-1.13.0
Question setv2
Judged at2026-10-16 10:39 UTC