Modelglass

AI model pricing and capability comparison from one honest, sourced registry

Developer ToolsLaunched Aug 2026

The story

Modelglass is a transparent pricing and comparison layer for AI models. It helps developers and AI-native teams compare models on cost, billing model, architecture, and capability across image, language, video, and audio categories. The platform tracks 34 image models across 17 providers, 69 language models across 25 providers, 30 video models across 11 providers, and 30 audio models across 18 providers, with regularly updated benchmark scores and pricing information.

Overview

Modelglass is a structured registry and comparison platform designed to solve the fragmentation problem in AI model information. It aggregates pricing, capabilities, and architectural details for image, language, video, and audio models from multiple providers into one searchable, sourced database. The platform serves developers, AI engineers, teams, researchers, and autonomous agents who need to evaluate and choose between AI models based on cost, context windows, benchmark performance, and provider availability. Users can compare models side-by-side, verify data provenance, track pricing changes over time, and integrate model data programmatically through an API and MCP server. Available as a web platform, iOS app, and VS Code extension.

Key features

Model Comparison

Compare AI models across cost, billing model, architecture, and capability metrics side-by-side from a unified registry.

Pricing Tracking

Monitor how model pricing and availability change over time across different providers and billing structures.

Benchmark Scores

Access tracked benchmark scores for image, language, and video models with context and provenance information.

API and MCP Integration

Query current model pricing and capability data programmatically through REST API and MCP server endpoints.

Multi-Platform Access

Access the registry through web, iOS app, and VS Code extension for pricing lookups and cost-aware model routing.

Sourced Registry

Verify where data came from and when it was last checked with transparent sourcing and update timestamps.

Use cases

1
Developers
Compare API pricing, context windows, capabilities, and provider availability when selecting models for applications.
2
AI Engineers
Access structured model metadata through API and MCP server for programmatic model selection and integration.
3
Research Teams
Inspect benchmark scores and their provenance to evaluate model performance across different evaluation frameworks.
4
Autonomous Agents
Query current model pricing and capability data to enable cost-aware and capability-aware model routing decisions.

FAQ

What models does Modelglass track?

Modelglass tracks 34 image models across 17 providers, 69 language models across 25 providers, 30 video models across 11 providers, and 30 audio models across 18 providers.

How often is the pricing and capability data updated?

The registry is regularly updated with the latest pricing and capability information. Specific update timestamps are provided for each model category.

How can I integrate Modelglass into my workflow?

You can use the REST API, MCP server, web platform, iOS app, or VS Code extension to access and integrate model comparison data.

What makes Modelglass different from other model comparison tools?

Modelglass focuses on sourced, structured data with transparent provenance, tracks pricing changes over time, and supports multiple billing models and architectures in one honest registry.

Who is Modelglass built for?

Modelglass is built for developers, AI engineers, teams, researchers, and agents who need to compare and choose between AI models based on cost and capability.