# om-ai-lab/OmAgent

[EMNLP-2024] Build multimodal language agents for fast prototype and production

Repository: https://github.com/om-ai-lab/OmAgent
Canonical: https://ross.abutalabs.com/products/omagent
Homepage: https://om-agent.com
Language: Python
License: Apache-2.0
License Family: permissive
Topics: large-language-models, multimodal-agent, vision-and-language, agent, workflow, smart-hardware, chatbot, gpt4, language-agent, llm, multimodal, rag, vlm, gpt, gradio, llama, llava, openai, python, gemini
Last push: 2025-03-19T11:36:13+00:00

## Health v2 (maintenance only)
Score: 31/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 12, release rhythm 40, longevity 56
- inputs: {"age_days": 790, "days_push": 532, "days_rel": 555, "gap_med": 20, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2665, forks 292 (observed 2026-08-28T04:07:09.050705+00:00)

## What it is
OmAgent is a Python library for building multimodal language agents, wrapping worker orchestration, task queues, and graph-based workflow orchestration behind a simple interface. It natively supports VLM models, video processing, mobile device connections, and agent algorithms like ReAct and CoT, with both distributed and Lite deployment modes.

## Use cases
- build a multimodal agent that reasons over video
- prototype a language agent with ReAct or CoT reasoning
- connect an agent to a mobile device for smart hardware control
- run a video understanding pipeline with a VLM
- deploy LLM agents locally with Ollama or LocalAI
- orchestrate a graph-based agent workflow with memory

## When to choose
- you need multimodal inputs like image, video, or audio in your agent
- you want a lightweight agent framework without heavy middleware overhead
- you need distributed scaling or a middleware-free Lite mode
- you want reusable agent components and graph-based workflows in Python

## When to avoid
- you only need simple text-only chatbot responses
- you prefer a mature ecosystem with extensive third-party integrations
- your project is not Python-based
- you need a no-code agent builder

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, rag, llm-inference, video-processing, workflow-automation, chatbot
- domain: artificial-intelligence, large-language-models, computer-vision
- platform: python, cross-platform
- tags: multimodal-agents, vlm, video-understanding, workflow-orchestration, llm-agents, distributed, ai-agents, multimodal, natural-language-processing

## Member repositories
- om-ai-lab/OmAgent (main) score 31

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:09.050705+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 2026-08-30T02:18:17.544733+00:00, confidence not recorded.
  - readme: https://github.com/om-ai-lab/OmAgent (fetched 2026-08-28T04:07:09.050705+00:00, sha 4c22ad323402)
  - homepage: https://om-agent.com (fetched 2026-08-29T10:00:59.015511+00:00, sha bcf7182a99cb)
- Data as of 2026-08-30T08:39:29.467469+00:00.