Guides
Set up your app, find a model, compare options and fix errors. Prices and model lists on these pages are read from the live catalog.
Roleplay apps
- Connect SillyTavern to TokenReplyChat Completion with a custom OpenAI-compatible source, one key for many models.
- JanitorAI proxy setup with TokenReplyAdd a proxy configuration with the full chat endpoint, your key and a model ID.
Models and free access
- Free AI API: models you can call without payingThe models that are free right now, read live from the catalog.
- Free Grok API accessCall the Grok models that are currently free through an OpenAI-compatible API.
- DeepSeek API guide: models, access and quick startCall DeepSeek models with the OpenAI SDK and compare the live options.
- Image generation APIGenerate images through the images endpoint or the chat endpoint.
Comparisons
- TokenReply vs OpenRouterFees, free models and API formats compared with sources.
- OpenRouter alternativesHosted gateways, direct provider APIs and self-hosted routers compared.
- Cheapest Claude models on TokenReplyLive Claude prices, ordered by input price, with cached input rates.
Tools and references
- IntegrationsSetup notes for frameworks, SDKs, coding agents and automation tools.
- Works withEvery app and tool with a TokenReply setup guide.
- API error referenceWhat common API errors mean and how to fix them.
- LLM price calculatorEstimate request and monthly costs from live prices.
- Price rankingsModels ordered by their current input, output or per-request price.
- Models by taskModels grouped by the endpoint they support, such as image generation.
Google
Anthropic
DeepSeek
Meta
Nvidia
Z.ai
OpenAI
XiaomiMiMo
MiniMax
Moonshot
Qwen
xAI
AionLabs
OpenCode Zen
Poolside
Tencent
Cohere
Cursor
dots studio
InclusionAI
Liquid AI
LongCat
Mistral
Stepfun