About llmdb.app

Who We Are

llmdb.app is an independent AI discovery platform built and maintained by the engineering team at Araho Digital. We are AI practitioners ourselves — we build with LLMs, ship AI-native products, and understand the challenges developers face when choosing models, writing prompts, and staying current with a field that changes weekly.

This site exists because we needed it. We were tired of hunting across docs, API pricing pages, and benchmark leaderboards to make simple decisions. llmdb.app puts the answers in one place, backed by transparent methodology and editorial standards you can verify.

Editorial Standards & Methodology

Every piece of content on llmdb.app follows a defined production standard:

• Model specifications(context windows, pricing, benchmarks) are sourced from official provider documentation and API references. We link to the source. When a provider has not published a figure, we label it clearly as “Undisclosed” or “Estimated” rather than fabricating data.
• Benchmark scoresare normalized to 0–100% scales from official provider releases or, where unavailable, from reproducible third-party evaluations. Each benchmark entry carries a provenance tag: “verified” (from provider docs), “estimated” (from community benchmarks), or “unscored” (no data available).
• Tool directory entries are researched and written by our editorial team. Affiliate relationships are clearly disclosed on each page. We do not accept payment for placement or ratings.
• News digests are compiled from primary sources (arXiv, company blogs, official announcements) and summarized by our editorial desk. Every digest links to its original source.
• Blog posts and comparisons are fact-checked against provider documentation before publication. Publication dates and last-updated dates are displayed on every article.

What This Platform Does

• AI Model Registry: Track context windows, token pricing, benchmark scores, and price history across 70+ frontier and open-source models — all sourced from provider docs with provenance tags.
• Live Prompt Sandbox: Test system prompts across multiple LLM providers in real time without signing up for individual API keys.
• Tool Directory: A curated catalog of 500+ AI applications categorized by technical utility, with transparent affiliate disclosure on every page.
• Daily AI News Digest: Human-curated summaries of significant developments in machine learning, model releases, and AI policy.

How This Site Is Built

llmdb.app runs on Next.js 15 (App Router) with TypeScript, Supabase for data storage, and Tailwind CSS for styling. Model data is synchronized from provider APIs and documentation. The site is deployed on Railway via Nixpacks with ISR (Incremental Static Regeneration) for fast page loads. We do not use AI to generate articles — every blog post, news digest, and directory entry is written by a human.

Contact & Corrections

If you spot an error in a model specification, benchmark score, or pricing entry, we want to fix it. Reach our editorial team at hello@araho.app or through our Contact Page. We respond to corrections within 48 hours.

For tool submissions, partnership inquiries, or press, use the same channels. We review every submission personally.