buildagentic

Tracker

Which agent frameworks are people actually using?

GitHub stars are a weak proxy for adoption and the only one every project publishes. Ranked lowest to highest, coloured by what each framework is for, so the shape of the field is visible before any single name is.
Every dot is a row in Agent engineering frameworks, with the source that reported it and the moment it was read.18 tracked →
GitHub stars
35 days of readings

LangChain leads on GitHub stars with 147k, 5.1x the median of 29k.

HighestLangChain · 147kMedianSemantic Kernel · 29kLowestGuardrails · 7kTracked18

Dot height = GitHub stars, dot colour = what the framework is for. Left to right is rank, lowest first — the horizontal position carries no other meaning.

  • orchestration6
  • multi-agent3
  • evaluation2
  • rag2
  • structured-output2
  • memory1
  • optimisation1
  • validation1

memory, optimisation, validation share the neutral swatch — the palette carries six distinct colours. Point at a mark to name it.

037k74k110k147kGitHub starsLowestHighestGuardrails7kSemantic Kernel29kLangChain147kGuardrails7kvalidationSynced from source, 1h agoInstructor14kstructured-outputSynced from source, 1h agoRagas16kevaluationSynced from source, 1h agoDeepEval18kevaluationSynced from source, 1h agoPydantic AI20kstructured-outputSynced from source, 1h agoLetta25kmemorySynced from source, 1h agoHaystack27kragSynced from source, 1h agoVercel AI SDK27korchestrationSynced from source, 1h agoMastra28korchestrationSynced from source, 1h agoSemantic Kernel29korchestrationSynced from source, 1h agosmolagents30kmulti-agentSynced from source, 1h agoDSPy38koptimisationSynced from source, 1h agoAgno42korchestrationSynced from source, 1h agoLangGraph42korchestrationSynced from source, 1h agoLlamaIndex52kragSynced from source, 1h agoCrewAI59kmulti-agentSynced from source, 1h agoAutoGen61kmulti-agentSynced from source, 1h agoLangChain147korchestrationSynced from source, 1h ago
Agent engineering frameworks →

History

What has moved since we started watching

35 distinct days of readings so far. Every observation is kept; none is overwritten.

  • LangChain

    147k +2k over 35d

    LangChain: 42 readings, 145k to 147k.

  • LangGraph

    42k +2k over 35d

    LangGraph: 42 readings, 40k to 42k.

  • CrewAI

    59k +2k over 35d

    CrewAI: 42 readings, 58k to 59k.

  • Mastra

    28k +941 over 35d

    Mastra: 42 readings, 27k to 28k.

  • DSPy

    38k +818 over 35d

    DSPy: 42 readings, 38k to 38k.

  • Pydantic AI

    20k +739 over 35d

    Pydantic AI: 41 readings, 19k to 20k.

  • DeepEval

    18k +649 over 35d

    DeepEval: 41 readings, 18k to 18k.

  • Vercel AI SDK

    27k +608 over 35d

    Vercel AI SDK: 42 readings, 26k to 27k.

  • AutoGen

    61k +585 over 35d

    AutoGen: 42 readings, 61k to 61k.

  • smolagents

    30k +553 over 35d

    smolagents: 41 readings, 29k to 30k.

  • Letta

    25k +508 over 35d

    Letta: 41 readings, 24k to 25k.

  • LlamaIndex

    52k +485 over 35d

    LlamaIndex: 42 readings, 52k to 52k.

  • Agno

    42k +470 over 35d

    Agno: 41 readings, 42k to 42k.

  • Ragas

    16k +407 over 35d

    Ragas: 40 readings, 15k to 16k.

  • Haystack

    27k +311 over 35d

    Haystack: 42 readings, 26k to 27k.

  • Instructor

    14k +176 over 35d

    Instructor: 41 readings, 14k to 14k.

  • Guardrails

    7k +141 over 35d

    Guardrails: 42 readings, 7k to 7k.

  • Semantic Kernel

    29k +117 over 35d

    Semantic Kernel: 42 readings, 28k to 29k.

More trackers

Other questions