⭕ LLMRing Registry

Warning: This is a pre-release registry. The pricing, limits, and capabilities listed here are still being validated and may be inaccurate. Do not use these numbers for production decisions. Always verify with official provider documentation.

The official model registry for LLMRing - providing up-to-date pricing, capabilities, and metadata for all major LLM providers.

Schema Version 3.5 - Dictionary-based models with standardized fields (e.g., max_input_tokens, dollars_per_million_tokens_input) and per-provider version numbers.

3
Providers
50+
Models
Daily
Updates

📁 Registry Files

📋 manifest.json - Registry metadata and version info 🟢 openai/models.json - OpenAI models and pricing 🔴 anthropic/models.json - Anthropic models and pricing 🔵 google/models.json - Google models and pricing 📊 models/summary.json - Summary statistics

🚀 Quick Start

Fetch Registry Data

// JavaScript
fetch('https://llmring.github.io/registry/manifest.json')
  .then(res => res.json())
  .then(data => console.log(data));

// Python
import requests
manifest = requests.get('https://llmring.github.io/registry/manifest.json').json()
print(manifest)

Get Model Information

// Fetch OpenAI models (v3.5 format)
fetch('https://llmring.github.io/registry/openai/models.json')
  .then(res => res.json())
  .then(data => {
    console.log(`Found ${Object.keys(data.models).length} OpenAI models`);
    console.log('GPT-4o pricing:', data.models['openai:gpt-4o']);
  });

🧑‍⚖️ Manual Curation Workflow

# 0) (Optional) Gather sources
uv run registry sources
uv run registry fetch-html --provider openai --output-dir html_cache
uv run playwright install chromium
uv run registry fetch --provider openai --output-dir pdfs

# 1) Generate a draft (best-effort) from PDFs (writes to drafts/)
uv run registry extract --provider openai --pdfs-dir pdfs

# 2) Review differences vs current curated file
uv run registry review-draft --provider openai --draft drafts/openai.2025-08-20.json
# Examine the .diff.json report next to the draft

# 3) Optionally accept all and generate a reviewed file
uv run registry review-draft --provider openai --draft drafts/openai.2025-08-20.json --accept-all

# 4) Promote the reviewed file to publish & archive
uv run registry promote --provider openai --reviewed drafts/openai.reviewed.json

📚 Documentation

For more information about the LLMRing project:

🔄 Update Schedule

The registry is automatically updated daily at 6 AM UTC via GitHub Actions. Each update fetches the latest pricing and model information directly from provider documentation.

📝 Data Schema

Each model entry includes:

Built with ❤️ by the LLMRing team | Report an Issue | Last updated: checking...