head to head · open source

Dify vs how-claude-code-works

Dify has 160,251 GitHub stars, 25,336 forks, 1,067 open issues and last shipped yesterday. how-claude-code-works has 3,800 stars, 714 forks, 1 open issues and last shipped 3 months ago. Dify leads on adoption by 4,117% (160,251 vs 3,800 stars). Dify is written in Python under a custom or non-standard licence; how-claude-code-works is written in various languages under MIT. Dify has attracted 16% as many forks as stars, how-claude-code-works 19%. Dify was the more recently maintained of the two, and both are self-hostable with no licence fee. The two share 3 topic tags (ai, claude, llm), so they are genuine substitutes rather than adjacent tools.

Two open source projects, one decision. Both are free and self-hostable — the differences are community size, license terms, language stack and release pace.

Dify ★ 160K how-claude-code-works ★ 3.8K category AI & Machine Learning

← all 20902 open source comparisons

Side by side

Dify how-claude-code-works
GitHub stars ★ 160K ★ 3.8K
License Custom / other MIT
Written in Python —
Last push 2026-11-02 2026-08-17
Forks ⑂ 25K ⑂ 714
Self-hosting Yes Yes
Data ownership Your server Your server

pick Dify if

  • You weight community size — 160K stars and counting
  • You want the Custom / other license terms
  • Your stack matches Python
  • You value the larger contributor base for long-term maintenance

full Dify profile →

pick how-claude-code-works if

  • You want the how-claude-code-works feature set and don't need the biggest community
  • You prefer the MIT license terms
  • You evaluated both and how-claude-code-works fits your workflow better

full how-claude-code-works profile →

About Dify

Dify Cloud · Self hosting · Documentation · Dify edition overview Dify is an open source LLM app development platform. Its intuitive interface combines AI workflow, RAG pipeline, agent capabilities, model management, observability features (including Opik, Langfuse, and Arize Phoenix) and more, letting you quickly go from prototype to production. Here's a list of the core features: Quick start Before installing Dify, make sure your machine meets the following minimum system requirements: CPU = 2 Core RAM = 4 GiB The easiest way to start the Dify server is through Docker Compose. Before running Dify with the…

read the full Dify overview →

About how-claude-code-works

How Claude Code Works Claude Code 架构学习笔记 — Agentic Coding System 深度分析 📘 在线阅读文档 English 🛠️ 想动手造一个?配套项目 Claude Code From Scratch — 4300 行 TypeScript / Python 两个版本,13 章分步教程,clean room 教育实现,从零造一个受 Claude Code 启发的 Coding Agent ⚖️ 声明 / Disclaimer :本项目是面向开发者的教育性架构分析。所有内容均为独立研究与推理,不代表 Anthropic 官方设计,也不保证与 Claude Code 真实内部实现一致。"Claude Code" 是 Anthropic 的商标,本项目与 Anthropic 无任何关联,亦不对外分发任何来自 Anthropic 的源码。 ⚖️ 写在前面 :2026.7.4: 距离3月底的源码泄露事件,到现在claude code又更新了许多新的功能,尤其是/goal /loop组成的 loop engineering, 以及dynamic workflow为代表的新的agent编排、观测和干预方法,这些新的功能的原理分析其实对于我们使用claude code帮助很大,因此我们近期也在持续逆向解析这些新的功能,不断更新该repo Claude Code 是目前使用最广泛的…

read the full how-claude-code-works overview →

More in AI & Machine Learning

Related comparisons

More AI Development Platforms projects

Compare either of these against the rest of the AI Development Platforms field.

Frequently asked questions

Is Dify or how-claude-code-works more popular?

Dify has 160,251 GitHub stars and how-claude-code-works has 3,800. Dify has the larger community by that measure.

Are Dify and how-claude-code-works free?

Both are open source. Dify has no licence declared in this registry, and how-claude-code-works is licensed under MIT. Both are free to self-host.

What is the difference between Dify and how-claude-code-works?

Dify is written in Python and how-claude-code-works in various languages. The practical differences are community size, licence terms, language stack and release cadence — all compared in the table above.

Which should I choose, Dify or how-claude-code-works?

Choose Dify if you want the larger community (160,251 stars) or its Custom / other licence terms. Choose how-claude-code-works if its feature set, stack or MIT licence fits better. Both are self-hostable.