MetaGPT
🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming
MetaGPT scores 5.9/10 (Solid) in Agents.
Language Python License MIT Stars 69.2k★ Type heavyweight Self-host Yes Repo FoundationAgents/MetaGPT Homepage https://www.deepwisdom.ai/
Quality
Solid 5.9
Adoption 71
Activity 11
Maturity 88
Community 75
Capability 86
Show the math
Overall 5.9/10
value = 5.9/10
score = value
Adoption 7.4/10
value = 71/100
score = 1 + 9 * value / 100
Activity 2/10
value = 11/100
score = 1 + 9 * value / 100
Maturity 8.9/10
value = 87.9/100
score = 1 + 9 * value / 100
Community 7.7/10
value = 74.7/100
score = 1 + 9 * value / 100
Capability 8.7/10
value = 85.9/100
score = 1 + 9 * value / 100
Key metrics
69.2k★ GitHub stars
0/90d Recent commits
0 devs/90d Recent contributors
HumanEval 85.9% HumanEval Pass@1 MetaGPT (with GPT-4) reaches 85.9% Pass@1 on HumanEval (and 87.7% on MBPP), reported as SOTA at publication. This is a code-generation capability benchmark rather than GAIA, but it is the primary published task-success number for this framework. Source: MetaGPT paper, arXiv 2308.00352 (ICLR 2024 Oral), Table 1. https://arxiv.org/html/2308.00352v6
3y Project age
5mo ago Last commit
Gotchas
No gotchas documented yet. Contribute one if you know a constraint we missed.
Labels
- Self-hosted
Self-hosted
You deploy and operate it yourself; there is no hosted option here.
- Heavyweight
Heavyweight
Resource-heavy to run (GPU, large memory, or multi-service stack).
multi-agent roles software-company framework agentic coding-agent