AI Model Directory

High-capability Claude family used for complex coding, reasoning, and long-context work.

Official source

Anthropic

Best for

Claude Code, difficult refactors, architecture review, and careful code reasoning.

Risk note

Relays should be checked for native Anthropic endpoint behavior and model substitution risk.

Balanced Claude family for daily coding assistance, agents, and editor workflows.

Official source

Anthropic

Best for

Claude Code daily use, Cursor/Cline style agent workflows, and cost-sensitive coding tasks.

Risk note

Check whether the relay supports your exact Claude Code configuration path before topping up.

OpenAI GPT and Codex-compatible workflows for CLI tools, SDKs, and coding assistants.

Official source

OpenAI

Best for

Codex CLI, OpenAI SDK, LiteLLM, Cursor, Cline, and OpenAI-compatible API workflows.

Risk note

Normalize token pricing and failed-request billing before comparing relay prices.

Google model family for long-context, multimodal, and cost-sensitive agent workloads.

Official source

Google

Best for

Long context, multimodal prompts, agent experiments, and fallback routes.

Risk note

Model naming varies across gateways; verify the exact model ID accepted by your tool.

Cost-efficient reasoning and coding model family frequently exposed through OpenAI-compatible APIs.

Official source

DeepSeek

Best for

Low-cost coding, reasoning, batch jobs, and OpenAI-compatible local toolchains.

Risk note

Good relay support does not mean Claude Code replacement; match it to OpenAI-compatible workflows.

Alibaba model family with strong Chinese, coding, and open-model ecosystem coverage.

Official source

Alibaba

Best for

Chinese-language workloads, low-cost coding, embeddings, and domestic model platform comparisons.

Risk note

Check whether the provider exposes the model through stable OpenAI-compatible naming.