# aiming-lab/Agent0

[ICML'26 & COLM'26] Agent0 Series: Self-Evolving Agents from Zero Data

Repository: https://github.com/aiming-lab/Agent0
Canonical: https://ross.abutalabs.com/products/agent0
Homepage: https://aiming-lab.github.io/Agent0
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-07-10T07:54:20+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 91, release rhythm 35, longevity 22
- inputs: {"age_days": 317, "days_push": 54, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1256, forks 150 (observed 2026-08-28T04:04:08.838231+00:00)

## What it is
Agent0 Series is a research framework for training self-evolving LLM agents from zero external data via tool-integrated reasoning and co-evolution between a curriculum agent and an executor agent. It includes Agent0 for text reasoning and Agent0-VL for multimodal vision-language reasoning with solver-verifier self-repair.

## Use cases
- train an llm agent without human-curated training data
- self-improve a language agent on math reasoning benchmarks
- evolve agents via curriculum and executor co-evolution
- build tool-integrated reasoning agents
- train a vision-language agent that self-evaluates and self-repairs
- reproduce research on autonomous agent evolution

## When to choose
- you want to train agents with zero external training data
- you need tool-integrated multi-turn reasoning agents
- you are doing research on self-improving or co-evolving agents
- you need multimodal (vision-language) agent self-evolution

## When to avoid
- you need a production-ready agent framework with managed infrastructure
- you lack GPU resources for LLM training
- you want a no-code or plug-and-play agent builder
- you need small-model or CPU-only training

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-training, machine-learning, prompt-engineering
- domain: artificial-intelligence, large-language-models, machine-learning
- platform: python, cross-platform
- tags: self-evolving-agents, zero-data, tool-integrated-reasoning, curriculum-learning, vision-language, research-code, co-evolution, ai-agents, research, linux, gpu

## Member repositories
- aiming-lab/Agent0 (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:08.838231+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 2026-08-30T05:07:15.485874+00:00, confidence not recorded.
  - readme: https://github.com/aiming-lab/Agent0 (fetched 2026-08-28T04:04:08.838231+00:00, sha d8583ec065ef)
  - homepage: https://aiming-lab.github.io/Agent0 (fetched 2026-08-29T12:17:49.993670+00:00, sha 7127fd6cd65d)
- Data as of 2026-08-30T08:39:29.467469+00:00.