python

DeepEval

DeepEval is an open-source evaluation framework from Confident AI. Every figure below carries the source that reported it and the moment it was observed; values that failed validation are withheld rather than shown.

What we know

15tracked fields

Every one carries a source and a date. Nothing on this page is a bare number.

Synced from sourceSynced from a source
12Current23h ago
Seeded, unreviewedWritten by us, unreviewed
3
Distinct sources
3
Last sync 1mo ago — succeeded.
Stars
18,403GitHub REST API
Contributors
337GitHub REST API
Licence
Apache-2.0GitHub REST API
Last sync
1mo agosucceeded

Every field

What we track, and where it came from

One row per field. The source link goes to the upstream project; the date is when that value was last observed, not when this page was built.

Every tracked field for DeepEval, with its source, its date, and where it stands.
FieldValueSourceAs ofStanding
ArchivednoSynced from sourceGitHub REST APICurrent23h ago
CategoryevaluationSeeded, unreviewedSeeded, unreviewedwritten 1mo ago
Contributors337Synced from sourceGitHub REST APICurrent23h ago
Forks1,968Synced from sourceGitHub REST APICurrent23h ago
LanguagepythonSeeded, unreviewedSeeded, unreviewedwritten 1mo ago
Last commit2026-09-22Synced from sourceGitHub REST APICurrent23h ago
LicenceApache-2.0Synced from sourceGitHub REST APICurrent23h ago
MaintainerConfident AISeeded, unreviewedSeeded, unreviewedwritten 1mo ago
Open issues642Synced from sourceGitHub REST APICurrent23h ago
PyPI packagedeepevalSynced from sourcePyPI registry metadataCurrent23h ago
PyPI released2026-09-22Synced from sourcePyPI registry metadataCurrent23h ago
PyPI version4.2.5Synced from sourcePyPI registry metadataCurrent23h ago
Repositoryhttps://github.com/confident-ai/deepevalSynced from sourceGitHub REST APICurrent23h ago
Requires Python<4.0,>=3.9Synced from sourcePyPI registry metadataCurrent23h ago
Stars18,403Synced from sourceGitHub REST APICurrent23h ago

History

How these numbers moved

Built from the observation log rather than from a current-state row. A mutable cell could not produce this at any price, and no backup would reconstruct it.

StarsCurrent23h ago
36 readings · +576

Line height = stars at each reading. Not to scale between fields.

Stars: 36 readings from 17,827 on 2026-08-24 to 18,403 on 2026-09-23.

17,827 18,403

ForksCurrent23h ago
36 readings · +123

Line height = forks at each reading. Not to scale between fields.

Forks: 36 readings from 1,845 on 2026-08-24 to 1,968 on 2026-09-23.

1,845 1,968

ContributorsCurrent23h ago
36 readings · +2

Line height = contributors at each reading. Not to scale between fields.

Contributors: 36 readings from 335 on 2026-08-24 to 337 on 2026-09-23.

335 337

Elsewhere

Related

Hiring

Open roles mentioning this

Matched from the job board by name. A mention is not a requirement — most listings name several tools, and which of them a role actually needs is recorded nowhere.

No open roles mention DeepEval. That is a fact about the roles we track, not about how widely it is used — the board covers agent-engineering roles at the companies we follow, and most tools never appear in one.

derived · matched by framework name · no fresher than the older of the listing and its classification

Compare DeepEval

Side by side, every tracked field, each figure still carrying the source that reported it and when.


Every figure on this site resolves to the source that reported it, the moment it was observed, and how it was measured. A value that fails validation is withheld rather than shown — withheld means we refused to publish it, not that it is zero.