Directory

Agent engineering frameworks

This directory tracks 18 open-source agent-engineering frameworks across GitHub activity, package downloads and release cadence. Every figure carries the source that reported it and the moment it was observed. Values that fail validation are withheld rather than shown, and the last verified snapshot stays on the page until a clean sync replaces it.
Frameworks
18
Stars tracked
682,608
With download data
10/18
Last sync
just nowquarantined
AgnoStars42,325+14GitHub REST API · just nowLettaStars24,864+13GitHub REST API · just nowDeepEvalStars18,419+16GitHub REST API · just nowRagasStars15,834+5GitHub REST API · just nowPydantic AIStars20,133+9GitHub REST API · just nowsmolagentsStars29,463+9GitHub REST API · just nowMastraStars28,295+25GitHub REST API · just nowVercel AI SDKStars26,922+20GitHub REST API · just nowGuardrailsStars7,443+1GitHub REST API · just nowInstructorStars13,939+4GitHub REST API · just nowDSPyStars38,243+25GitHub REST API · just nowHaystackStars26,584+4GitHub REST API · just now
2 agent engineering frameworks, sorted by stars. Each row shows the last time the ranked figure was observed.
Select to compareFrameworkStarsDownloads/wkLast commitLanguageLicenceOpen roles7-day starsChecked
MastraMastra28,2951,161,907typescriptunknown47-day stars: 7 readings, 28.1K to 28.3K.Currentjust now
Vercel AI SDKVercel26,92217,883,737typescriptunknown187-day stars: 7 readings, 26.8K to 26.9K.Currentjust now

1 of 3 frameworks selected — pick one more to compare

CompareClear

Sources: GitHub REST API, npm registry downloads, PyPI registry metadata, pypistats.org downloads. Every value carries the moment it was observed; figures that failed validation are withheld rather than shown. SLA: current within 48h, ageing to 168h, stale beyond 336h.

indexable · canonical /frameworks?language=typescript

Also tracked

Other datasets

Same engine, same provenance rules — every figure carries the source that reported it and when.

Smoke dataset

A fixture dataset that exercises the ingestion, validation and freshness machinery.

4 tracked

Language and embedding models

Language, embedding and speech models tracked across Hugging Face Hub download activity and OpenRouter context windows and pricing.

18 tracked

Local inference hardware

GPUs, accelerators and SoCs for running models locally, with VRAM, memory bandwidth, power draw and list price.

10 tracked