Agentes de IA
WorkArena
WorkArena: ServiceNow's benchmark for how well web agents handle everyday knowledge-work tasks.
★ 275 estrelasNOASSERTION
Ver o guia →LibreAI / Agentes de IA / AgentBench
// agents · THUDM/AgentBench
AgentBench
AgentBench é um projeto de agentes de IA de código aberto de THUDM (THUDM/AgentBench no GitHub). Tem 3.765 estrelas no GitHub e 288 forks, é publicado sob a licença Apache-2.0 e foi atualizado pela última vez no GitHub em 8 de fevereiro de 2026.
A rigorous benchmark for LLMs acting as agents.
★ 3,8 mil estrelas⑂ 288 forksApache-2.0atualizado fev 2026
▦ visual: o README deste projeto não inclui nenhuma captura incorporável,
então aqui vai um marcador. O guia abaixo está completo mesmo assim.
então aqui vai um marcador. O guia abaixo está completo mesmo assim.
// dados do projeto
Dados do projeto
- Repositório
- THUDM/AgentBench
- Estrelas no GitHub
- 3.765
- Forks
- 288
- Licença
- Apache-2.0
- Última atualização no GitHub
- 8 de fevereiro de 2026
- Categoria
- Agentes de IA
- Tipo de ficha
- Guia de instalação completo
// como funciona
Como funciona
- Start with a use case
Pick one concrete task so you can judge whether the project fits.
- Fetch the source
Clone the repo and check the prerequisites listed in the README.
- Set up the environment
Install what it needs and configure credentials or model access.
- Try the sample workflow
Run a minimal example end to end before wiring it into your stack.
// recursos
Por que as pessoas gostam tanto
Agentic evaluation
Tests models as actors in environments.
ICLR 2024
Peer-reviewed benchmark design.
Multi-environment
Diverse tasks beyond static QA.
THUDM research
From a leading academic lab.
Python harness
Runnable evaluation code.
Apache-2.0
Permissive licensing.
Advertisement
// execute você mesmo
Execute você mesmo
- Step 1 — Clone the repo
Download the source and move into the project folder.
git clone https://github.com/THUDM/AgentBench.git cd AgentBench - Step 2 — Install
Set up a virtual environment and install dependencies.
python -m venv .venv && source .venv/bin/activate pip install -r requirements.txt - Step 3 — Run tasks
Evaluate a model on a task subset per the docs.
# see README for the task runners
// casos de uso
O que você pode criar
- Evaluate a new model as an autonomous agent.
- Compare agent frameworks on standard tasks.
- Reproduce ICLR 2024 agent results.
- Design better agentic evals for your own models.
Advertisement
// perguntas frequentes
Perguntas frequentes
Who maintains AgentBench?
It is maintained by THUDM at THUDM/AgentBench, with 288 forks.
Is AgentBench free to use?
Yes. It is released under the Apache-2.0 license, which permits commercial use, modification, and redistribution. Read the LICENSE file for the exact terms.
Do I need a paid API to use it?
You need model access to evaluate; the harness itself is free.
How does it compare to SWE-bench?
AgentBench covers diverse agentic environments; SWE-bench targets real-world software engineering specifically.
How do I get started?
Follow the repo docs to run a task subset against your model.
// continue explorando
Ferramentas semelhantes
Agentes de IA
sglang
High-performance serving for the LLMs that power your agents.
★ 36,9 mil estrelasApache-2.0
Ver o guia →Agentes de IA
OSWorld
OSWorld — benchmark multimodal agents on open-ended tasks in real computer environments.
★ 3,2 mil estrelasApache-2.0
Ver o guia →Agentes de IA
deepteam
Open-source red-teaming framework that stress-tests LLMs and AI agents.
★ 3,0 mil estrelasApache-2.0
Ver o guia →Agentes de IA
AI-fundamentals
Chinese-language course on AI basics: GPU architecture, CUDA, LLMs, and agents.
★ 2,9 mil estrelasApache-2.0
Ver o guia →Agentes de IA
ADR
Uber's ADR: discover, observe, benchmark, and detect threats across enterprise AI agents, deployed at Uber.
★ 1,9 mil estrelasApache-2.0
Ver o guia →// alternativas
Alternativas em Agentes de IA
Os outros projetos de Agentes de IA com mais estrelas no catálogo do LibreAI — a lista realista de alternativas se AgentBench 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 AgentBench: continue explorando dos dois lados dele.
- vocode-core★ 3,8 mil
Build voice-based LLM agents with a modular, open-source toolkit.
- morphik-core★ 3,7 mil
An open-source multimodal retrieval engine for agent pipelines.
- atmosphere★ 3,8 mil
A portable AI agent runtime for the JVM — write one @Agent, run it on any Java AI stack.
- antigravity-sdk-python★ 3,7 mil
Google's Python SDK for building AI agents on the Antigravity platform and Gemini models.
- SimpleMem★ 3,8 mil
Research-backed lifelong memory for LLM agents, text and multimodal.
- goclaw★ 3,6 mil
OpenClaw rebuilt in Go — multi-tenant AI agent teams with 5-layer security in a single binary.
Baixe o AgentBench
Grátis e aberto: pegue o código-fonte, dê uma estrela no repositório e comece a criar.