github-featured/microsoft--agent-lightning
2
0
Fork 0
Code Issues Pull Requests Deployments Models Agent Notes Packages Projects Releases Wiki Activity

Docs/update readme stable docs link (#544)

Browse Source 
Co-authored-by: GitHub Actions <actions@github.com>
This commit is contained in:
Zhiyuan He
2026-08-17 15:02:48 +08:00
committed by GitHub
parent 5e9c711724
commit 2914f3e2ef
1 changed files with 16 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+16 -16
View File
@@ -5,14 +5,14 @@
<p align="center"><em>3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses!</em></p>
<p align="center">
<a href="docs/">Documentation</a> &nbsp;·&nbsp; Technical Report (Coming Soon) &nbsp;·&nbsp; <a href="LICENSE">MIT License</a>
<a href="https://microsoft.github.io/agent-lightning/stable/">Documentation</a> &nbsp;·&nbsp; Technical Report (Coming Soon) &nbsp;·&nbsp; <a href="LICENSE">MIT License</a>
</p>
> Agent Lightning was completely refactored in v1.0. For legacy releases earlier than v1.0, see [this branch](https://github.com/microsoft/agent-lightning/tree/v0.x).
## ⚡ Key Features
- 🪶 **~3,500 lines of core Python:** We treat simplicity as the first principle.
- 🪶 **~3,500 lines of code:** We treat simplicity as the first principle.
- 🧩 **Train with real agent harnesses:** Agents interact with the model through the Agent Lightning v1.0 proxy with **ZERO changes**, while keeping tools, context, control flow, and environments in the loop.
- ☸️ **Native Kubernetes support:** Run agents directly as Kubernetes Jobs without relying on external sandbox services.
- 💻 **Full coding agent training example:** Using only **6K training samples**, an end-to-end Qwen3.5-9B workflow improves SWE-bench Verified from **41.8% to 56.4%**, a gain of **14.6 percentage points**. We release the full pipeline, including data cleaning, reward-hacking prevention, and training scripts.
@@ -27,7 +27,7 @@ uv sync
bash scripts/setup_verl.sh 0.8.0 cu130
```
See the [Installation Guide](docs/1-installation.md) for details.
See the [Installation Guide](https://microsoft.github.io/agent-lightning/stable/1-installation/) for details.
## ⚡ Architecture
@@ -56,24 +56,24 @@ We evaluate Agent Lightning v1.0 across several practical training domains, incl
| Section | Content |
|---------|---------|
| [Installation](docs/1-installation.md) | Base environment and `verl` GPU stack |
| [Quick Start](docs/2-quick-start.md) | Local first run and end-to-end flow |
| [Basics](docs/3-basics.md) | Components, rollouts, events, and trajectories |
| [Trainer Configuration](docs/4-trainer-configuration.md) | `verl` integration and trace aggregation |
| [API Gateway Configuration](docs/5-api-gateway-configuration.md) | Gateway and model proxy settings |
| [Controller Configuration](docs/6-controller-configuration.md) | Local and Kubernetes runners |
| [Asynchronous Training](docs/7-asynchronous-training.md) | Collocated async collection and pause/drain |
| [Installation](https://microsoft.github.io/agent-lightning/stable/1-installation/) | Base environment and `verl` GPU stack |
| [Quick Start](https://microsoft.github.io/agent-lightning/stable/2-quick-start/) | Local first run and end-to-end flow |
| [Basics](https://microsoft.github.io/agent-lightning/stable/3-basics/) | Components, rollouts, events, and trajectories |
| [Trainer Configuration](https://microsoft.github.io/agent-lightning/stable/4-trainer-configuration/) | `verl` integration and trace aggregation |
| [API Gateway Configuration](https://microsoft.github.io/agent-lightning/stable/5-api-gateway-configuration/) | Gateway and model proxy settings |
| [Controller Configuration](https://microsoft.github.io/agent-lightning/stable/6-controller-configuration/) | Local and Kubernetes runners |
| [Asynchronous Training](https://microsoft.github.io/agent-lightning/stable/7-asynchronous-training/) | Collocated async collection and pause/drain |
## ⚡ Examples
| Example | Description |
|---|---|
| [Calc-X](docs/8-example-calc-x.md) | POC math reasoning example with AutoGen and MCP calculator tools, requiring only one GPU. |
| [GSM8K](docs/9-example-gsm8k.md) | POC grade-school math reasoning example. |
| [ScienceWorld](docs/10-example-science-world.md) | Interactive science tasks in a text-based environment. |
| [Search-R1](docs/11-example-search-r1.md) | Multi-turn retrieval and reasoning agent. |
| [LLM-in-Sandbox](docs/12-example-llm-in-sandbox.md) | General agent with computer and code execution tools. |
| [Coding Agent](docs/13-example-coding-agent.md) | Coding agent trained with repository tests. |
| [Calc-X](https://microsoft.github.io/agent-lightning/stable/8-example-calc-x/) | POC math reasoning example with AutoGen and MCP calculator tools, requiring only one GPU. |
| [GSM8K](https://microsoft.github.io/agent-lightning/stable/9-example-gsm8k/) | POC grade-school math reasoning example. |
| [ScienceWorld](https://microsoft.github.io/agent-lightning/stable/10-example-science-world/) | Interactive science tasks in a text-based environment. |
| [Search-R1](https://microsoft.github.io/agent-lightning/stable/11-example-search-r1/) | Multi-turn retrieval and reasoning agent. |
| [LLM-in-Sandbox](https://microsoft.github.io/agent-lightning/stable/12-example-llm-in-sandbox/) | General agent with computer and code execution tools. |
| [Coding Agent](https://microsoft.github.io/agent-lightning/stable/13-example-coding-agent/) | Coding agent trained with repository tests. |
## ⚡ Articles