# gabber-dev/gabber

Build AI applications that can see, hear, and speak using your screens, microphones, and cameras as inputs.

Repository: https://github.com/gabber-dev/gabber
Canonical: https://ross.abutalabs.com/products/gabber
Homepage: https://gabber.dev
Language: Python
License: NOASSERTION
License Family: other
Last push: 2025-12-20T23:29:00+00:00

## Health v2 (maintenance only)
Score: 44/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 58, release rhythm 35, longevity 28
- inputs: {"age_days": 404, "days_push": 256, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1111, forks 106 (observed 2026-08-28T04:03:37.742196+00:00)

## What it is
Gabber is an open-source engine for building real-time multimodal AI applications that can see, hear, and speak, using graph-based orchestration of nodes for media ingestion, transcription, LLM/VLM inference, and TTS. It includes a visual app builder, supports multi-participant multi-stream apps via WebRTC (LiveKit), and can run local LLMs with llama.cpp or vLLM.

## Use cases
- build a realtime voice chat assistant with my microphone and camera
- create an AI app that watches my screen and answers questions
- build a personal trainer app that counts reps from video
- orchestrate STT, LLM, and TTS pipelines with sub-second latency
- build a multi-participant realtime AI companion
- visually compose AI workflows from nodes without writing glue code
- run a local LLM-powered voice agent on my own machine

## When to choose
- you need sub-second realtime multimodal AI (voice, video, screen) orchestration
- you want a graph/visual builder for composing AI pipelines
- you want self-hosted realtime AI with local LLM support

## When to avoid
- you only need simple batch LLM inference or a chatbot without realtime media
- you need a fully managed cloud service rather than self-hosting LiveKit and the engine
- your project is a lightweight script where a graph engine is overkill

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, speech-recognition, tts, computer-vision, streaming, workflow-automation, web-framework
- domain: artificial-intelligence, large-language-models, speech-processing, computer-vision, developer-tools
- platform: python, self-hosted
- tags: realtime-ai, multimodal, graph-based-orchestration, webrtc, livekit, voice-agent, visual-builder, vlm, real-time, docker, macos, linux, web-server

## Member repositories
- gabber-dev/gabber (main) score 44

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:37.742196+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:43:23.290848+00:00, confidence not recorded.
  - readme: https://github.com/gabber-dev/gabber (fetched 2026-08-28T04:03:37.742196+00:00, sha c602c6990cc9)
  - homepage: https://gabber.dev (fetched 2026-08-29T12:47:02.747332+00:00, sha 26b1412820d9)
  - site_page: https://docs.gabber.dev (fetched 2026-08-29T12:47:02.756410+00:00, sha 58b81f6cfd80)
- Data as of 2026-08-30T08:39:29.467469+00:00.