// agents · comet-ml/opik

opik

opik é um projeto de agentes de IA de código aberto de comet-ml (comet-ml/opik no GitHub). Tem 22.451 estrelas no GitHub e 1.849 forks, é publicado sob a licença Apache-2.0 e foi atualizado pela última vez no GitHub em 8 de outubro de 2026.

Open-source LLM observability: trace, evaluate and monitor apps and agents

★ 22,5 mil estrelas⑂ 1,8 mil forksApache-2.0atualizado out 2026

Ver no GitHub ↗Official site ↗

▦ visual: o README deste projeto não inclui nenhuma captura incorporável,
então aqui vai um marcador. O guia abaixo está completo mesmo assim.

// dados do projeto

Dados do projeto

Repositório: comet-ml/opik
Estrelas no GitHub: 22.451
Forks: 1.849
Licença: Apache-2.0
Última atualização no GitHub: 8 de outubro de 2026
Categoria: Agentes de IA
Tipo de ficha: Guia de instalação completo
Site oficial: Official site

// como funciona

Como funciona

Decide cloud or self-hosted
Use Comet's cloud for the fastest start, or self-host the full platform with Docker Compose or Helm.
Get Opik
Install the Python SDK with pip, or clone the repository to self-host the server.
Configure the SDK
Run opik configure, add the tracing decorator to your functions, and point the SDK at your workspace.
Trace your first call
Run your app once and open the Opik dashboard to inspect the full trace and score it.

// recursos

Por que as pessoas gostam tanto

End-to-end tracing

Logs LLM calls including nested function and agent traces, so you can replay exactly what happened.

Automated evaluations

Score outputs with automated evaluation methods and run experiments against datasets.

Production dashboards

Monitor quality, usage and feedback from live traffic with production-ready dashboards.

Agent and RAG observability

Built for agentic workflows and RAG systems, not just single completions.

Framework-agnostic

Works across ecosystems - the repo highlights LangChain and LlamaIndex among its integrations - without locking you into one framework.

Self-hostable

Deploy locally with Docker Compose or at scale with Kubernetes and Helm, or use the Comet cloud.

// execute você mesmo

Execute você mesmo

Clone the repository
Get the source code from GitHub and enter the project directory.
```
git clone https://github.com/comet-ml/opik.git
cd opik
```
Install the SDK and configure
Add Opik to your Python project and connect it to your workspace or local server.
```
pip install opik
opik configure
```
Or self-host the platform
Clone the repo and start the full Opik suite locally with the provided script (run inside the clone).
```
./opik.sh
```

// casos de uso

O que você pode criar

Debug a RAG pipeline by inspecting retrieval and generation traces
Run offline evaluations before shipping a prompt or model change
Monitor production answer quality and user feedback over time
Trace multi-step agent runs to find where a workflow went wrong

// perguntas frequentes

Perguntas frequentes

Who maintains this project?

It is maintained by Comet (the company behind Comet ML) as an open-source project on GitHub.

Is it free to use?

Yes - Opik is Apache-2.0 licensed, and the README stresses the full platform can be self-hosted free without a commercial license. Comet also offers a hosted cloud option.

Does it need a paid API key?

Opik itself needs no paid key when self-hosted; the LLM calls you trace still use your own provider keys, and the cloud option needs a Comet account.

How does it compare to alternatives?

Compared with Langfuse, Opik positions itself as a fully Apache-2.0, self-hostable platform with evaluation and optimization built in; the README includes a comparison table against alternatives.

How do I get started?

Os outros projetos de Agentes de IA com mais estrelas no catálogo do LibreAI — a lista realista de alternativas se opik não for o ideal para você.

openclaw★ 391,6 mil
A personal AI that really does things — on any OS, with your data staying yours.
superpowers★ 294,9 mil
An agentic skills framework & software development methodology that works.
ECC★ 275,2 mil
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
hermes-agent★ 252,0 mil
The agent that grows with you — a Python framework for adaptive AI agents.
deepseek-harness★ 245,5 mil
DeepSeek's plugin-based agent harness where everything is a plugin.

// mais em agents

Mais em Agentes de IA

Projetos de Agentes de IA próximos no catálogo, os mais próximos em estrelas de opik: continue explorando dos dois lados dele.

NemoClaw★ 22,7 mil
NVIDIA's reference stack for running AI agents more securely inside OpenShell sandboxes
babyagi★ 22,4 mil
Yohei Nakajima's autonomous task agent - now a self-building agent framework in Python
n8n-mcp★ 23,0 mil
An MCP server that lets Claude, Cursor and Windsurf build n8n workflows for you
generative_agents★ 22,2 mil
Stanford's Generative Agents simulation - 25 AI characters living in a virtual town
go-micro★ 23,1 mil
A framework for building agents and services
Swarm★ 22,1 mil
Swarm — OpenAI's experimental framework for lightweight multi-agent orchestration with agents and handoffs, MIT

Baixe o opik

Grátis e aberto: pegue o código-fonte, dê uma estrela no repositório e comece a criar.

Ver no GitHub ↗← Mais ferramentas de Agentes de IA