Model List and Price Comparison

See the details of all models across all your relay sites at a glance. Supports cross-site price comparison, automatic selection of the best group, and batch availability verification to help you make the most cost-effective choice.

Features at a Glance

💰 Real-time Price Comparison: Automatically calculates the actual unit price of each model at different sites based on billing ratios, group ratios, and exchange rates.
🎯 Best Path Recommendation: In the "All Accounts" view, the system automatically highlights the "Lowest Price" across all sites or the "Optimal Group" for the current site.
🔍 Multi-dimensional Deep Filtering: Supports combined filtering by site source, API credentials, providers (OpenAI, Anthropic, etc.), billing modes, and tags.
✅ Batch Availability Verification: Supports model connectivity testing, token compatibility checks, and CLI proxy availability verification.
📊 Billing Transparency: Clearly distinguishes between "Token Billing" and "Per-Call Billing," and provides official ratio comparisons.

How to Access

Click the "Model List" icon at the bottom left of the extension popup.
Or go to Settings → Model Management.

Core Operating Guide

1. Select Data Source

In the selector at the top of the page, you can switch between different data sources:

All Accounts: Aggregates all site accounts you've added to enable cross-site comparison.
Specific Account: View the full model catalog of a specific site.
API credentials: View models supported by keys saved in the API Credential Library (no site account required).

2. Intelligent Price Comparison

Model cards display detailed billing information for the model:

Billing Ratio: Shows the site's model ratio (e.g., 1.5) and completion ratio (e.g., 2).
Group Ratio: Shows the additional ratio for your current group (e.g., default, VIP).
Actual Unit Price: The system automatically converts prices to Per 1M Tokens (USD/CNY) or Per Call.
Lowest Price Badge: In the "All Accounts" view, sites with the best price will have a green badge.

3. Group Switching and Simulation

If a site supports multiple user groups (e.g., default, vip, svip), you can switch groups in the filter bar to preview real-time price changes for different tiers.

💡 Tip: The system defaults to selecting the most cost-effective available group for the model.

4. Model Verification

Click the verification icons on the right side of the model card:

Verify Model: Sends a lightweight request to confirm if the model is currently truly available.
Verify CLI Compatibility: Tests if the model supports streaming output and can be normally called by command-line tools.
Batch Verification: Click "Batch Verify" in the toolbar to queue tests for all currently filtered models. You can also manually select multiple models via checkboxes to perform verification only on the selected set.

List Control Options

Option	Description
Search Box	Supports fuzzy search by Model ID (e.g., `gpt-4o`) or site name.
Billing Mode	Filter models by "Token Billing" or "Per-Call Billing."
Provider Filter	Quickly lock onto models from OpenAI, Anthropic, Google, Meta, and other vendors.
Sort By	Supports sorting by Name, Price (Low to High), and Account Count.
Display Settings	Control whether to show CNY prices, show ratio columns, and show endpoint types.

FAQ

Question	Solution
Why do some models show a price of 0?	The site may have set this model to be free, or the site hasn't provided valid billing data, in which case it falls back to the default catalog.
How is the exchange rate calculated for cross-site comparison?	The extension has a built-in fixed exchange rate of 1 USD = 7.3 CNY (referencing the New API convention) to provide a uniform comparison baseline.
What does "Optimal Group" mean?	When a model is available in multiple groups (e.g., `default` and `vip`), the extension automatically selects the group with the lowest ratio to calculate the price.
Why does the verification result show "Unknown"?	Please check if your API Key has permission to call the model, or if the site is currently triggering a rate limit.

API Credential Library: How to save Base URL + API Key pairs without accounts.
Usage Analytics: View the actual spending generated by these models.
Supported Sites: View the architecture types that support automatic price recognition.