Compatibility
MarkdownThe API formats Ailoot supports, the models available in each, and the clients that use them.
Full guide
Ailoot implements the public contracts of three APIs. If an application supports OpenAI, Anthropic, or Gemini and lets you set a custom URL, it can also work with Ailoot. Change the URL and key to connect.
All paths are relative to https://api.ailoot.ai. The SDK URL column shows the base URL to configure: https://api.ailoot.ai/v1 (with /v1) or https://api.ailoot.ai (without /v1), because SDKs append the rest of the path themselves.
| Format | Path | SDK URL | ID |
|---|---|---|---|
| OpenAI Chat Completions | POST /v1/chat/completions | with /v1 | 139 |
| OpenAI Responses | POST /v1/responses | with /v1 | 106 |
| Anthropic Messages | POST /v1/messages | without /v1 | 150 |
| Gemini | POST /v1beta/models/{m}:generateContentPOST /v1beta/models/{m}:streamGenerateContent | without /v1beta | 29 |
| OpenAI Embeddings | POST /v1/embeddings | with /v1 | 6 |
| OpenAI Images | POST /v1/images/generationsPOST /v1/images/edits | with /v1 | 30 |
| OpenAI Audio | POST /v1/audio/speechPOST /v1/audio/transcriptions | with /v1 | 20 |
| Video and music | POST /v1/videos/generationsPOST /v1/music/generations | with /v1 | 18 |
| Decisions (Jev) | POST /v1/systemone | no SDK; use the full URL | 1 |
| Model list (OpenAI, Anthropic) | GET /v1/models | with /v1 | all |
| Model list (Gemini) | GET /v1beta/models | without /v1beta | all |
The catalog contains 1256 model cards and 2692 documented methods. Alternative caller IDs or aliases are listed under the selected method when available.
Models by API format
Most text models—Claude, GPT, DeepSeek, Qwen, Kimi, GLM, MiniMax, and Grok—are available in three formats: OpenAI Chat, OpenAI Responses, and Anthropic Messages. Gemini models are available in Anthropic Messages and native Gemini format. NVIDIA models primarily use OpenAI Chat.
OpenAI Chat Completions
claude-sonnet-5-5, gpt-6.1-sol, gemini-3.6-flash, grok-4.7, deepseek-v4-pro, qwen3.8-max, k3, glm-5.3, MiniMax-M3, nvidia/nemotron-3-ultra-550b-a55b, meta/muse-glimmer-30b, poolside/laguna-xs-2.1, claude-opus-5-5, gpt-6-astra and another 51
OpenAI Responses
claude-sonnet-5-5, gpt-6.1-sol, grok-4.7, deepseek-v4-pro, qwen3.8-max, k3, glm-5.3, MiniMax-M3, claude-opus-5-5, gpt-6-astra, grok-4.7-build-fast, deepseek-v4.1-flash, qwen3.8-flash, kimi-for-coding and another 36
Anthropic Messages
claude-sonnet-5-5, gpt-6.1-sol, gemini-3.8-flash, grok-4.7, deepseek-v4-pro, qwen3.8-max, k3, glm-5.3, MiniMax-M3, nvidia/nemotron-3-ultra-550b-a55b, meta/muse-glimmer-30b, poolside/laguna-xs-2.1, claude-opus-5-5, gpt-6-astra and another 71
Gemini
gemini-3.8-flash, gemini-3.1-pro, gemini-pro-agent, gemini-3.8-flash-tiered, gemini-3.8-flash-medium, gemini-3.8-flash-low, gemini-3.8-flash-high, gemini-3.7-flash, gemini-3.7-flash-tiered, gemini-3.7-flash-medium, gemini-3.7-flash-low, gemini-3.7-flash-high, gemini-3.6-flash, gemini-3.6-flash-tiered and another 13
For the full list with prices and filters, see the model catalog.
Which formats clients use
Below are applications from our integration catalog, grouped by the format they use. Many support several formats.
OpenAI Chat Completions
Hermes Agent, OpenClaw, opencode, Qwen Code, Kimi CLI, Crush, Aider, Goose, Factory Droid, Amp, Cline, Roo Code, Kilo Code, Continue, Zed, VS Code + Copilot Chat, JetBrains AI Assistant, Xcode, Open WebUI, LibreChat, Cherry Studio, Chatbox, LobeChat, Jan, SillyTavern, OpenAI SDK (Python), OpenAI SDK (Node.js), Vercel AI SDK, LangChain, n8n, Dify, AIChat, AnythingLLM, avante.nvim, big-AGI, BoltAI, codecompanion.nvim, Continue CLI, CrewAI, Flowise and 30 more
OpenAI Responses
OpenAI Codex CLI, Hermes Agent, OpenClaw, Kimi CLI, Factory Droid, Amp, Zed, VS Code + Copilot Chat, Cherry Studio, OpenAI SDK (Python), OpenAI SDK (Node.js), Vercel AI SDK, LangChain, n8n, Kilo Code CLI, LLM, Mastra, OpenAI Agents SDK, Pydantic AI
Anthropic Messages
Claude Code, Hermes Agent, OpenClaw, Qwen Code, Kimi CLI, Crush, Goose, Factory Droid, Amp, Cline, Roo Code, Zed, VS Code + Copilot Chat, LibreChat, Cherry Studio, LobeChat, Jan, Anthropic SDK (Python), Vercel AI SDK, LangChain, n8n, AIChat, avante.nvim, big-AGI, codecompanion.nvim, CrewAI, GitHub Copilot CLI, gptel, Kilo Code CLI, NextChat, Pi, Trae Agent, Trae IDE
Gemini
OpenClaw, Qwen Code, Gemini CLI, Kimi CLI, Amp, Vercel AI SDK, n8n, AIChat, Google GenAI SDK (Python), gptel, Pi
Supported features within each format
| Feature | How it works |
|---|---|
| Streaming responses | SSE in all text formats: stream: true or :streamGenerateContent?alt=sse. Billed text is delivered in one block after the cost is calculated, rather than word by word. See details |
| Tools (tools, function calling) | Yes, if the model supports them. Tool descriptions follow the selected API. Gemini text requests also accept googleSearch and urlContext, separately or together, but not alongside functionDeclarations in the same request. Native image output supports only client-side functionDeclarations |
| Image inputs | Yes, for models that can process images. The estimated image cost is included in the reserve. In Gemini format, use PNG, JPEG, or WebP in the inlineData field of the user message |
| Structured output (JSON) | response_format in OpenAI, and its equivalents in Anthropic, if supported by the model. In Gemini format, responseMimeType in generationConfig accepts text/plain or application/json; use responseSchema or responseJsonSchema supported by the exact model. Capabilities and parameter combinations depend on the model |
| Reasoning | reasoning_effort / reasoning in OpenAI, thinking in Anthropic, thinkingConfig in Gemini, for reasoning models |
| Token counting | Separate token counting (/v1/messages/count_tokens) is a separate method; its availability is checked for the selected model and key. Messages support alone does not confirm support for token counting |