> ## Documentation Index
> Fetch the complete documentation index at: https://docs.aihubmix.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Model details and comparison

> Read model capabilities, context limits, performance, and pricing units. Check the exact version and ID, then compare candidates on the same task.

Select a model's name to open its details and check whether it fits your inputs, outputs, and integration. Preserve the complete name and version suffix when copying its model ID.

<h2 id="read-details">
  What to check
</h2>

| Information | How to use it |
| - | - |
| Input / output types | Check which text, image, audio, or video inputs are accepted and what the model returns |
| Context / maximum output | Check the context capacity and the per-generation output limit separately, in tokens |
| Capabilities and parameters | Verify requirements such as tool calling, structured output, reasoning, and supported parameters |
| Model version | Confirm the exact version, free or discounted variant, and availability |
| TTFT / throughput | Compare time to first token and output speed; actual results vary by task |

A `-` means that a metric is unavailable. It does not mean zero latency or no capability.

<h2 id="pricing">
  Check pricing
</h2>

Review input, output, and additional capability charges separately. Text is commonly priced per million tokens. Images and video may be priced per image, per second, or by generation settings. Check the conditions for tiered prices, caching, and promotions.

A **Free** label does not make every additional capability free or guarantee unlimited concurrency. Use the current pricing and limits in the model's details.

<h2 id="compare-and-try">
  Compare and try candidates
</h2>

1. Shortlist models in the catalog and use the comparison entry to review their differences.
2. Compare context, capabilities, and prices using equivalent units.
3. Try the same input on each candidate and review quality, latency, and actual usage.
4. Copy the chosen ID and use Playground or the [quickstart](/en/quick-start) to call the API.

If a model is retired or unavailable, choose an available version and check parameter compatibility. To preserve the model name used in your application, see [model mapping and fallback](/en/api/Model-Mapping-Fallback).

Related: [Choose a model](/en/guides/choose-models) · [Smart search](/en/guides/smart-search) · [Models API](/en/api/Models-API)

Last updated: 2026-10-13


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.