Alternatives
Products that do what xmllm – Structured LLM streaming output using lenient XML parsing does
Hi HN. I made a little JS library for streaming structured data from LLMs using leniently-parsed XML as a medium. E.g. await simple('fun pet names', { schema: { name: Array(String) }, model: 'openrouter:mistralai/ministral-3b' }); // => ["Daisy", "Whiskers", "Rocky"] Demos: xmllm.j11y.io When using LLMs, I've ended up gravitating towards boring time-tested XML-esque tag-based delimiters instead of JSON/function-calling for the following reasons: - Diverse presence in training corpuses (consider flavours of content commonly adjacent to these syntaxes vs. JSON) - HTML was…
- 1
- 2
- 3
- 4SE
I built a CLI tool in Go that extracts structured data (JSON, CSV, Parquet) from messy PDFs and HTML pages. The core idea: LLMs are great at understanding structure but wasteful for bulk data extraction. So smelt uses a two-pass architecture: 1. A fast Go capture layer parses the document and detects table-like regions 2. Those regions (not the whole document) get sent to Claude for schema inference — column names, types, nesting 3. The Go layer then does deterministic extraction using the inferred schema This means the LLM is never in the hot path of actual data processing. It figures out…
Mar 2026 · github.com
- 5LS
Ever had json.loads() explode halfway through an LLM stream? langdiff fixes that with a schema + callback approach. Define your schema → attach callbacks → push streaming tokens → get structured events immediately.
2025 · github.com
- 6CR
hi everyone. how does moving llm call prompts and output structure definitions away from code into configuration land sound? would you use something like this if it was stable and well documented enough? please don't hold back the criticism. i appreciate all feedback (constructive & otherwise).
2024 · github.com
- 7AO
I've built an airgapped Retrieval-Augmented Generation (RAG) system for question-answering on documents, running entirely offline with local inference. Using Llama 3, Mistral, and Gemini, this setup allows secure, private NLP on your own machine. Perfect for researchers, data scientists, and developers who need to process sensitive data without cloud dependencies. Built with Llama C++, LangChain, and Streamlit, it supports quantized models and provides a sleek UI for document processing. Check it out, contribute, or suggest new features!
2024 · github.com
- 8ET
We built a browser extension (Chrome + Firefox) that captures the runtime DOM and exports it as JSON. Not the pre-render source (HTML/CSS/JS, templates, bundles) and not a screenshot — but the live, post-render state the browser is actually displaying: - visibility/hidden, disabled/required - current input values and validation/validationMessage - dataset attributes - trimmed text - stable selector paths Why: LLMs often miss or guess UI state. Screenshots are too opaque, pre-render source is too noisy. A structured snapshot gives reproducible context for debugging…
2025
- 9IM
Live demo here: http://fonctionlabs.com:8000 Similarly to aka_sh (guess we were working parallelly on similar topics), I created with my brother a chainlit-based webapp, which summarizes Youtube videos in order to gain time. It works as an RAG-based LLM, and is very light in the sense that it does not use RAG libraries like langchain or llamaindex. You can use it with your own OpenAI API key. It also supports local models like Mistral, or Llamma. It is ofc open-source, and you can deploy with Docker if you choose. Some of the next steps are: - using whisper to be able to compute a…
2024 · github.com
- 10FA
2021 · fifthtry.com
- 11AO
Hi, We are building an open-source framework for loading and structuring LLM context to create accurate and explainable LLM answers using knowledge graphs and vector stores. We built the tool with four main concepts in mind: 1. Loader -> uses dlt in the backend to load and structure the data 2. Cognify step -> creates a graph with summaries, labels and factoids that are interconnected across the documents and stored as a representation in the vector store 3. Optimizer -> Uses DSPy to optimize LLM queries, and we plan to extend it to most of the knobs we can turn, like chunking etc. 4. Search…
2024 · github.com
- 12LF
LLM from URL —— A free AI chat completion service directly from URL Usage: In the address bar of any web browser, type your question after https://818233.xyz/ and hit Enter to get the instant answer. You know the best part of this? Whitespace in the url is supported in most web browsers! You can also use curl or Wget to retrieve the appended url by replacing any whitespace with a '+' character. If you need to have an actual '+' character in your question, just use '++'. Example: The url "https://818233.xyz/hi there" in any web browser will return the same answer…
2025 · 818233.xyz
- 13TA
Hi HN! This Christmas I made `tysm`, a rust crates which simplifies using ChatGPT with Structured Outputs. Essentially you just tell it the rust type you want the output to conform to, and it handles creating the JSON Schema, passing it to ChatGPT's chat-completions API, and deserializing the response. Structured Outputs are very useful because they guarantee that the response will conform to whatever type you expect. I'm already finding my crate very useful since I love using Structured Outputs but I hate writing JSON Schemas myself. Technical details for rust heads: To generate the JSON…
2024 · github.com
- 14LT
I wanted to share a project I've been working on for the past few weeks: llgtrt. It's a Rust implementation of a HTTP REST server for hosting Large Language Models using llguidance library for constrained output with NVIDIA TensorRT-LLM. The server is compatible with the OpenAI REST API and supports structured JSON schema enforcement as well as full context-free grammars (via Guidance). It's similar in spirit to the Python-based TensorRT-LLM OpenAI server example but written entirely in Rust and built with constraints in mind. No Triton Inference Server involved. This also serves as a demo…
2024 · github.com
- 15UA
Hey HN! After using a combination of Unsloth and Axolotl a lot, and finding it generally painful to figure out the right performance tuning for things like batch sizing and multi-GPU sharding, I wrote a small Python lib that sets up known-good LoRA training configurations for Llama 3.1 8B and 70B Instruct, and includes helpers for distilling from larger models or training on serverless finetuning platforms, and includes a walkthrough for distilling DeepSeek-R1 into a Llama 3.1 8B LoRA... But you can use it for pretty much any finetuning task, not just distilling large models!
2025 · github.com
- 16A1
I've seen a lot of comments about how complex frameworks like LangChain can be. Over the holidays, I wanted to see how minimal an LLM framework could get if we stripped away everything non-essential. The result is an LLM framework in just 100 lines of code. These 100 lines capture what I see as the core abstraction of most LLM frameworks: a nested directed graph that breaks down tasks into multiple LLM steps, with branching and recursion to enable agent-like decision-making. From there, you can layer on more advanced features like agents, RAG, task decomposition, and more. I’ve intentionally…
2025 · github.com
- 17LO
Introducing LLM Optimize, a toy proof-of-concept library for LLM-guided blackbox optimization using GPT-4. Perform optimization on problems beyond the usual numerical methods, such as code-based AutoML and natural language rubric-based optimization. Check it out: https://github.com/sshh12/llm_optimize
2023 · github.com
- 18OS
2023 · github.com
- 19MC
Hi HN, I'm excited to introduce Mixlayer, a platform I've been working on over the past 6 months that allows you to code and deploy prompts using simple JavaScript functions. Mixlayer recreates the developer experience of using LLMs locally without having to do all of the local setup yourself. I originally came up with this idea when using LLMs on my MacBook and thought it’d be cool to build a product that makes it easy for everyone. It compiles your code to a WASM binary and runs it alongside a custom inference stack I wrote in Rust. When you integrate LLMs in this way, your code and the…
2024 · mixlayer.com
- 20TA
Hi HN, TamedTable is an LLM harness for data ETL. And yes, it was developed using AI, meaning you can take the entire specification and recreate it to your desires: https://github.com/ZSvedic/TamedTable
Aug 2026 · tamedtable.com
- 21CA
Hi HN, I've been working with LLMs in production for a while both as a solo dev building apps for clients and working at an AI startup. The one thing that always was a pain was to pay OpenAI/Gemini/Anthropic a few dollars a month just for me to say "test" or have a CI runner validate some UI code. So I built this server called ChunkBack, that mocks the popular llm provider's functionality but allows you to type in a deterministic language: `SAY "cheese"` or `TOOLCALL "tool_name" {} "tool response"` I've had to work in some test environments and give good results for experimenting…
Nov 2025 · github.com
- 22BL
Hello everyone! I am Jan, CTO and one of the creators of Pathway, the real-time data processing framework. I’m excited to share Pathway’s ready-to-use AI Pipelines, configurable with just YAML! These frameworks offer out-of-the-box solutions for AI search, RAG, and more—optimized for real-time indexing and in-memory processing. What makes it simple? YAML templates! The pipeline templates are fully customizable using YAMLs to fit your needs, from changing the data sources to the choice of the LLM model, all without touching Pathway’s Python code. Thanks to the Pathway data processing engine,…
2024 · pathway.com
- 23PA
Hey HN! We just launched PromptL: a templating language built to simplify writing complex prompts for LLMs like GPT-4 and Claude. Why PromptL? Creating dynamic prompts for LLMs can get tricky, even with standardized APIs that use lists of messages and settings. While these formats are consistent, building complex interactions with custom logic or branching paths can quickly become repetitive and hard to manage as prompts grow. PromptL steps in to make this simple. It allows you to define and manage LLM conversations in a readable, single-file format, with support for control flow and…
2024 · promptl.ai
- 24PL
Hey HN! We realised that LLMs are great at generating code for super popular libraries like React. But they kinda suck at using less popular/newly released libraries, forcing us to stick to established tools and hindering innovation. There is already a standard for creating documentation for LLMs (llmstxt.org), but in my experience the implementations have not been great so far. `llms.txt` works as a good index of the available pages, but in many cases they link to HTML pages. This is a waste for LLMs to parse through (For example, Hono's [best…
2025 · github.com
Ranked by how close each launch is in meaning, then by votes. Refine with a description →