Alternatives
Products that do what Unsiloed AI – #1 on olmOCR-Bench does
Most of the document parsers fail on real world challenges like complex tables, handwritten documents, historical document scans, equations, multi-column layouts, complex reading order, etc. We built Unsiloed Parser to handle exactly these cases. Our latest parser v3.1 achieved #1 rank and scored 88.0 strict pass-rate on olmOCR-Bench. We ran the evaluation across 1,403 PDFs and 8,413 unit tests using the unmodified upstream Allen AI scorer (olmocr==0.4.27) and found Unsiloed beats 18 other OCR services, including GPT-5.5, Claude Opus 4.7, LlamaParse, Reducto, Azure Document Intelligence, AWS…
- 1
- 2I4
It's our new text-to-image model: a 9.3B single-stream diffusion transformer trained entirely from scratch. We focused heavily on controllability through structured JSON prompts, with strong text rendering, spatial awareness through bounding box guidance, and color palette control. It has the best text rendering of any open-weight model we've tested so far, and the NF4 quantized checkpoint runs on a single 24GB GPU. For more technical details and examples see our blog post: https://ideogram.ai/blog/ideogram-4.0/ We will be happy to answer any questions :)
Jun 2026 · github.com
- 3
- 4
- 5OA
Hi HN, we're Kiran and Vijay! Over the past two years, we have built a columnar storage engine for observability: logs, metrics, and traces. Today, it's exciting for us to show what we've built on top of that foundation: LLM Agent Observability. Given how non-deterministic agents are, storing all traces without sampling was critical for us. But these traces tend to be in the MBs, sometimes GBs - we needed to store them inexpensively. We also needed the queries and analyses to be fast. To meet both these goals, we store them in S3 in our own parquet-like file format, and query them using AWS…
Jul 2026 · oodle.ai
- 6DA
I’m one of the co-founders of Doctly AI. I wanted to share our story. We didn’t originally set out to build a PDF-to-Markdown parser. It all started when we were building a RAG solution for a company that deals with regulatory agencies. All of their data was in PDFs, and as it is apparently with lawyers, they like to print and scan documents to make it hard on their counterparts. These documents contained complex tables that barely make sense, are rotated, and handwriting is mixed in between. Many pages are number ruled and potentially rotated. We spent a lot of time trying to get clean data…
2024
- 7
High performance storage engine for efficient LLM inference and GPU Training.
16h ago · theopenlake.com
- 8CB
I built a small benchmark to test CLI coding agents on blind bug detection. A challenger agent injects bugs and writes ground truth (`bugs.json`). A different reviewer agent audits the repo without seeing ground truth, and an LLM matcher scores bug-to-finding assignments. Current run: 50 repos, 150 challenges, 450 reviews, 2,603 injected bugs. Weighted detection: Claude 58.05%, Codex 37.84%, Gemini 27.81%. LLM-judge benchmarks are easy to get wrong, so I’d really appreciate critical feedback on benchmark fairness, scoring/matching methodology, and obvious failure modes I’m missing. Full…
Feb 2026 · github.com
- 9
- 10OW
This was not supposed to become a product. When PaddleOCR-VL-1.6 dropped, independent benchmarks put it at the top of document parsing models. I had to try it. I needed a provider, but there simply isn't one ready for production that I would trust. So i set one up myself. I assumed that even after getting it running, serving a vision-language model would be expensive. It turns out the opposite is true. Once I had it running properly, the cost was absurdly low. At proper GPU utilization, the cost is only around $1 per 1,000 pages. The nearest competitors are either much lower quality (Azure…
Jul 2026 · openparser.dev
- 11PA
Hi HN, I run a small ecommerce shop and, until recently, spent way too many evenings copy-pasting supplier invoices into Excel so my books stayed clean and my expense tracking was granular. It finally hit me that I’d rather invest that time in code than in Ctrl-C/Ctrl-V, so I built ParsePoint.app. Why I built it • Manual invoice entry was swallowing 4 hours every month. • Accounting software APIs felt heavyweight for my use-case; I really just wanted clean rows in a spreadsheet. • I love tinkering with AI models and needed an excuse to spin one up in production. What it does 1. Upload a…
2025 · parsepoint.app
- 12
Turn Unstructured Documents into Actionable Data with AI
Jun 2026 · sigixtract.com
- 13MD
We’re excited to share ML-Dev-Bench, a new open-source benchmark that tests AI agents on real-world ML development tasks. Unlike typical coding challenges or Kaggle-style competitions, our benchmark simulates end-to-end ML workflows including: - Dataset handling and preprocessing - Debugging model and code failures - Implementing new model architectures - Fine-tuning and improving existing models With 30 diverse tasks, ML-Dev-Bench evaluates agents across critical stages of ML development. To complement this, we built Calipers, a framework that provides systematic performance evaluation and…
2025 · github.com
- 14TT
Hey HN, this is deepan from trulytyped (https://trulytyped.com). I am building a document writing app which makes it extremely easy to figure out how a document was created. Now that any text can be AI generated, how do you tell if something was actually generated or composed. It is impossible to detect AI after a piece of text has been generated. No amount of watermarking, linguistic checks or vibe checks work consistently. The AI detectors that schools and journals use are easy to bypass. Why do we need to solve this problem - First of all, this is not an anti-AI stance. I have…
May 2026 · trulytyped.com
- 15AP
Hello Hacker News, I’m releasing TXT Blah Blah Blah Lite, an open-source plain-text AI reasoning engine powered by semantic embedding rotation. It generates 50 coherent, self-consistent answers within 60 seconds — no training, no external APIs, and zero network calls. Why this matters Six top AI models (ChatGPT, Grok, DeepSeek, Gemini, Perplexity, Kimi) independently gave it perfect 100/100 ratings. For context: Grok scores LangChain around 90 MemoryGPT scores about 92 Typical open-source LLM frameworks score 80-90 Key features Lightweight and portable: runs fully offline as a single…
2025 · github.com
- 16UI
Hey everyone! I am excited to share updates on four of my & my teams' open-source projects that take large-scale search systems to the next level: USearch, UForm, UCall, and StringZilla. These projects are designed to work seamlessly together, end-to-end—covering everything from indexing and AI to storage and networking. And yeah, they're optimized for x86 AVX2/512 and Arm NEON/SVE hardware. USearch [1]: Think of it as Meta FAISS on steroids. It's now quicker, supports clustering of any granularity, and offers multi-index lookups. Plus, it's got more native bindings than probably…
2023 · usearch-images.com
- 17OU
The traditional pipeline for unstructured data extraction typically follows these steps: 1. Image → OCR Model (e.g., Google Vision) → Layout Model (e.g. Surya) → LLM → Final Answer However, this can be streamlined using a Vision-Language Model (VLM): 2. Image → VLM → Final Answer Recently VLMs have improved a lot for OCR and document understanding tasks, specifically the Qwen-2.5-VL series. We can run the Qwen-2.5-VL-7B-AWQ model locally with just 16GB VRAM, and perform end-to-end information extraction (fields and table extraction) without any external models. Hallucination with VLMs One…
2025 · github.com
- 18DE
Hi HN! I built Docuglean, an open-source SDK for intelligent document processing that works with OpenAI, Mistral, Google Gemini, and Hugging Face models. The idea came from repeatedly writing boilerplate code to extract structured data from invoices, receipts, and other documents. Instead of wrestling with different API formats, I wanted a unified interface that: - Extracts structured data using Zod/Pydantic schemas - Classifies and splits multi-section documents (e.g., medical records) - Processes documents in batches with automatic error handling - Works locally without APIs (for…
Nov 2025 · github.com
- 19
1:1 layout math scanner. Export to native Word OMML & LaTeX.
May 2026
- 20BA
I built CodeLens.AI - a tool that compares how 6 top LLMs (GPT-5, Claude Opus 4.1, Claude Sonnet 4.5, Grok 4, Gemini 2.5 Pro, o3) handle your actual code tasks. How it works: - Upload code + describe task (refactoring, security review, architecture, etc.) - All 6 models run in parallel (~2-5 min) - See side-by-side comparison with AI judge scores - Community votes on winners (blind voting) - Each evaluation gets reflected in the overall AI model leaderboard, showing us best ones Why I built this: Existing benchmarks (HumanEval, SWE-Bench) don't reflect real-world developer tasks. I wanted to…
Oct 2025 · codelens.ai
- 21
Chat with PDFs using AI — Normal & OCR modes, fully offline.
Jun 2026 · edupulse666.gumroad.com
- 22WB
Hi HN, Our research team just released the best performing and most efficient reranker out there, and it's available now as an open weight model on HuggingFace. Reranker v2 was designed specifically for agentic RAG, supports instruction following (our v1 was the first to introduce this), and is multilingual. Along with this, we're also open source our eval set, which allows you to reproduce our benchmark results. By releasing these datasets, we are also advancing instruction-following reranking evaluation, where high-quality benchmarks are currently limited. Please give it a try and let us…
2025 · huggingface.co
- 23MM
Hi HN! We (Thomas and Stéphan, hello!) recently released Model2Vec, a Python library for distilling any sentence transformer into a small set of static embeddings. This makes inference with such a model up to 500x faster, and reduces model size by a factor of 15 (7.5M params or 15/30MB on disk, depending on whether you use float16 or float32). This allows you to embed 50-100k documents per second on a cpu on a macbook. This reduction of course comes at a cost: distilled models are worse than their parent models. Even so, they are actually a lot better than large sets of conventional…
2024 · github.com
- 24YA
Hi HN, I hacked together Yurl.ai last week—a bare-bones URL shortener in F# / .NET 9 using a simple CQRS + event-sourcing setup. The gimmick: for each new link, it grabs the page text and asks a small language model to suggest a short, readable slug (4–8 chars). I’m curious whether these model-made names beat the usual random hashes. Tech bits: Actors as aggregates SQLite for both event store and read model Source (MIT): https://github.com/OnurGumus/YURL Live demo: https://yurl.ai (free tier—please be gentle) Happy to hear what you think.
2025 · yurl.ai
Ranked by how close each launch is in meaning, then by votes. Refine with a description →