AI Model Collections
Browse curated lists of the models on AnonRouter: the best models for coding and roleplay, uncensored models, image generation APIs, long context models, the cheapest routes, and the ones that run inside hardware enclaves. Each list is built from what providers publish about the route and kept in step with the live catalog, and every model in them runs through one OpenAI-compatible API that never stores your prompts.
End-to-End Encrypted AI Models
Encrypted in your browser and opened only inside a verified enclave, so nothing on the path can read the prompt.
Uncensored AI Models
Models their own provider publishes as unmoderated, reached through an account that is not tied to your identity.
AI Models for Roleplay and Creative Writing
Models their provider describes for roleplay and storytelling, plus the unmoderated ones these setups usually need.
Best AI Models for Coding
Tuned for writing, reviewing, and refactoring code, with the context windows an agent loop actually needs.
Image Generation Models
Text to image and image editing routes, priced per image and callable with the same API key as everything else.
AI Reasoning Models
Models that run an explicit thinking pass before answering, with the effort level under your control.
AI Models with Tool Calling
Function calling for agents, MCP servers, and anything that needs the model to hand work back to your code.
AI Models with Vision
Multimodal models that read screenshots, scanned pages, and diagrams in the same request as the prompt.
Confidential Compute AI Models
Inference that runs inside attestable hardware enclaves, which makes the guarantee checkable rather than promised.
Zero Data Retention AI Models
Providers that keep no copy of the prompt or the response once the request is finished.
Long Context AI Models
200K tokens and up, for whole repositories, long transcripts, and document sets that will not fit anywhere else.
Cheapest AI Models
Text models at $0.15 per million input tokens or less, several of which still run inside enclaves.
AI Models with Web Search
Models that can pull live results into an answer without a separate search integration in front of them.
AI Models with Prompt Caching
Reuse a long system prompt or document across a session and pay a lower rate on every turn after the first.
Best Text-to-Speech Models
Speech synthesis with a published voice list, billed per character of input text.
Audio and Speech Models
Speech synthesis, transcription, music, and sound design, all behind the same key.
Text Embedding Models
Vector embeddings for search, retrieval, and clustering, priced per million tokens.
AI Models with Multi-Provider Failover
The same model served by several providers, with routing policy choosing between them per request.