Model Marketplace and Rankings

The Model Marketplace helps you discover available models and compare pricing. Rankings show real usage within the platform. These pages answer different questions and should not be treated as interchangeable.

Use the Model Marketplace

1. Open the Model Marketplace.
2. Search or filter to narrow the available models.
3. Review protocol labels, billing mode, and input/output pricing.
4. Open the detail view, then copy the exact model name for configuration.

Pricing views and units

• Standard / Recharge: the site exposes different pricing views. Use the current account page and console rules to determine actual billing.
• /1M: price per 1,000,000 tokens.
• /1K: price per 1,000 tokens.

Convert prices to the same unit before comparing them. A /1M figure cannot be compared directly with a /1K figure.

Input, output, and cache pricing

• Input pricing covers prompts, conversation history, and other context sent to a model.
• Output pricing covers content generated by the model and often differs from input pricing.
• Cache pricing may apply to reusable cached tokens for supported models. Do not assume a cache rate when none is displayed.

Use this estimate after normalizing every rate to the same token unit:

Estimated cost = input tokens × input rate
               + output tokens × output rate
               + cached tokens × cache rate

Interpret rankings

The Rankings page supports daily, weekly, monthly, and yearly views. It may show popular models, token volume, provider market share, and upward or downward trends.

• Token volume measures usage on this platform.
• Provider market share aggregates token volume by model provider.
• Trends describe recent usage changes, not absolute capability.

Popularity is not a quality benchmark

Choose a model using task fit, protocol compatibility, input/output cost, reliability, and your own evaluations. Ranking position alone is not enough.