lagent vs OxyGent
Python frameworks for LLM agents compared
lagent and OxyGent are Python frameworks for building applications with language models, agents, and tools. lagent emphasizes agent components, session memory, and synchronous or asynchronous workflows, while OxyGent focuses on reusable components and configurable multi-agent collaboration.

lagent: Build LLM Agents with a PyTorch-Inspired Framework
InternLM’s Lagent is a lightweight Python framework for building LLM-based agents, using a PyTorch-inspired design for composing agents and message passing. It supports synchronous and asynchronous workflows, tools, memory, and multi-agent applications.

OxyGent: Build Modular Multi-Agent Systems in Python
OxyGent is a Python framework that uses its Oxy abstraction to assemble tools, models, and agents into observable, evolvable multi-agent systems. It is aimed at developers building and running agent teams with flexible collaboration patterns.
| lagent | OxyGent | |
|---|---|---|
| Language | Python | Python |
| License | Apache-2.0 | — |
| Stars | 2.3k | 2.1k |
| Forks | 248 | 276 |
| Last analyzed | Oct 15, 2026 | Oct 29, 2026 |
Key differences
- lagent describes structured agent messages, session memory, and customizable message aggregation; OxyGent centers its tools, models, and agents on a shared Oxy component model.
- lagent supports workflows from a single tool-using agent to multi-agent setups; OxyGent is oriented toward multi-agent systems, including agent relationships and sub-agents.
- lagent highlights synchronous and asynchronous component variants; OxyGent describes dynamic and hybrid planning, dependency mapping, and visual debugging.
- OxyGent includes a web service entry point and evaluation engines; lagent’s listed features focus on agent components, memory, parsing, and action execution.
- lagent reports an Apache-2.0 license; OxyGent reports NOASSERTION, while its README advertises Apache 2.0, so verify the applicable license.
- Both are Python projects and neither is archived. lagent lists 2.3k stars and 245 forks; OxyGent lists 2.1k stars and 276 forks.
Choose lagent if you…
- need synchronous or asynchronous agent workflows with session memory.
- want to connect structured agent messages, output parsing, and tool execution.
- are building a tool-using assistant or a workflow that can range from one agent to several.
Choose OxyGent if you…
- want reusable, swappable components for multi-agent applications.
- need to define agent relationships or explore dynamic and hybrid planning.
- want a web service entry point and features for execution visibility or evaluation.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.