axflow/axflow

Health v2 · maintenance only

Adoption not part of the score

1121 stars · 41 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Axflow is a modular TypeScript framework for building natural language powered AI applications, composed of packages like @axflow/models (SDK with React hooks and streaming), axgen (connecting data to LLMs), and axeval (LLM output evaluation). Modules can be adopted incrementally or combined into an end-to-end AI development framework.

Use cases

• build chatbots and LLM-powered apps in TypeScript
• stream LLM responses to a React frontend
• connect my own data to large language models for RAG
• evaluate the quality of LLM outputs
• integrate OpenAI or other model providers with a zero-dependency SDK
• load and chunk documents for vector search

When to choose

• you want a code-first, modular TypeScript toolkit for LLM apps
• you need streaming utilities and React hooks for AI UIs
• you want to adopt RAG and evaluation tooling incrementally in one ecosystem

When to avoid

• you need a mature, battle-tested framework with a large community
• you work outside the TypeScript/JavaScript ecosystem
• you need the planned extract/serve/finetune modules, which are not yet released

Facets

framework · maturity experimental

llm-inference rag agent-framework sdk http-client web-framework large-language-models artificial-intelligence web-development developer-tools cross-platform llm streaming react-hooks llm-evaluation modular-sdk retrieval-augmented-generation nodejs typescript

Member repositories

For agents

markdown · JSON · MCP: product_card(name="axflow/axflow")