Show HN: A lightweight LLM proxy to get structured results from most LLMs

Hey HN! After struggling with complex prompt engineering and unreliable parsing, we built L1M, a simple API that lets you extract structured data from unstructured text and images. curl -X POST https://api.l1m.io/structured \ -H "Content-Type: application/json" \ -H "X-Provider-Url: demo" \ -H "X-Provider-Key: demo" \ -H "X-Provider-Model: demo" \ -d '{ "input": "A particularly severe crisis in 1907 led Congress to enact the Federal Reserve Act in 1913", "schema": { "type": "object", "properties": { "items": { "type": "array", "items": { "type": "object", "properties": { "name": { "type": "string" }, "price": { "type": "number" } } } } } } }' This is actually a component we unbundled from our larger because we think it's useful on its own. It's fully open source (MIT license) and you can: - Use with text or images - Bring your own model (OpenAI, Anthropic, or any compatible API) - Run locally with Ollama for privacy - Cache responses with customizable TTL The code is at https://github.com/inferablehq/l1m https://github.com/inferablehq/l1m with SDKs for Node.js, Python, and Go. Would love to hear if this solves a pain point for you!