Tag
5 posts tagged.
Promptfoo, DeepEval and Ragas are the three most-recommended open-source LLM evaluation frameworks. For a TypeScript team the first question is which of them you can actually run. We checked the registries in August 2026, and the usual answer is out of date.
A developer-lens 2026 comparison of the Vercel AI SDK and LangChain.js: a thin, edge-native model-and-UI toolkit versus a broader retrieval-and-agent orchestration framework, and why the @ai-sdk/langchain adapter means many Next.js teams should use both.
Qdrant, Weaviate, and Chroma are the three open-source vector databases you can actually self-host. We compare embedded versus server, self-host cost, license, managed pricing, and when pgvector wins.
Mastra, LangChain.js, and LlamaIndex.TS for building AI agents in TypeScript, 2026: stars, license, which is still maintained, and how to choose.
Hands-on June 2026 review of the LlamaIndex.TS RAG starter for Next.js. The starter still works, the project is officially deprecated, and the deploy-paths matrix below has three options that still ship. Honest cost numbers, vector store picks, and what to point a fresh codebase at instead.
