# 💬 About TikHub AI API # TikHub AI Proxy Documentation ## Overview TikHub AI Proxy offers unified access to multiple leading AI models through a single API key, featuring centralized billing, rate control, and usage monitoring. The service allows developers and teams to switch between AI providers seamlessly without changing application logic. > **TikHub AI Proxy is currently in BETA.** If you encounter any issues or would like to request additional model support, please contact us. --- ## Getting Started ### 1. Create an Account and Obtain an API Key 💡 Sign up at 👉 [https://ai.tikhub.io](https://ai.tikhub.io) Once registered, generate your API key from the console. ### 2. Review Supported Models Your TikHub API key grants access to all models listed on the pricing page: 💡 [https://ai.tikhub.io/pricing](https://ai.tikhub.io/pricing) Refer to this page for model availability, pricing, and usage limits. ### 3. Retrieve Your API Key Access your API key from the developer console: 💡 [https://ai.tikhub.io/console/token](https://ai.tikhub.io/console/token) Keep this key secure, as it is required for all API requests. --- ## Migrating Existing Integrations If you are already using official AI provider APIs (OpenAI, Claude, Gemini, etc.), migrating to TikHub AI Proxy requires **no code changes**. ### Migration Steps 1. Update the **base URL** to: ``` https://ai.tikhub.io ``` 2. Replace your existing API key with your TikHub API key 3. Keep all request formats unchanged > **Important SDK Compatibility Notice:** > TikHub AI Proxy is **NOT compatible with official SDKs** (OpenAI Python SDK, Anthropic SDK, Google AI SDK, etc.). You must use **direct HTTP/HTTPS requests** to interact with the AI Proxy API. --- ## New Integrations If you are not currently integrated with any official AI provider, please refer to the endpoint documentation below to begin using TikHub AI Proxy from scratch. --- ## Supported Platforms and Models TikHub AI Proxy currently exposes **76 models** across six providers, unified behind OpenAI-, Anthropic-, and Google-compatible endpoints. | Platform | Endpoint Style | Notes | |---|---|---| | **OpenAI** | `/v1/responses`, `/v1/chat/completions`, `/v1/images/generations` | GPT-4, GPT-4.1, GPT-4o, GPT-5 series, O1/O3/O4 reasoning models, image generation | | **Anthropic (Claude)** | `/v1/messages` (native format) | Opus, Sonnet, Haiku — including extended-thinking variants | | **DeepSeek** | `/v1/chat/completions` | Chat, Reasoner (R1), V3 series, OCR | | **Google (Gemini)** | `/v1beta/models/{model}:generateContent` | Text/chat and image-generation models | | **Moonshot (Kimi)** | `/v1/chat/completions` (OpenAI-compatible) | Chat completion | | **Alibaba (Qwen)** | `/v1/chat/completions` (OpenAI-compatible) | Chat completion | | **Video generation** | Dedicated video endpoints | Sora 2 / Sora 2 Pro, Veo, Kling, Seedance | For the full, always-current list with pricing and rate limits, call `GET /v1/models` or check the [pricing page](https://ai.tikhub.io/pricing). ### OpenAI **GPT-4 series:** `gpt-4-vision-preview` **GPT-4.1 series:** `gpt-4.1`, `gpt-4.1-2025-04-14`, `gpt-4.1-mini`, `gpt-4.1-mini-2025-04-14`, `gpt-4.1-nano`, `gpt-4.1-nano-2025-04-14` **GPT-4o series:** `gpt-4o`, `gpt-4o-2024-05-13`, `gpt-4o-2024-08-06`, `gpt-4o-2024-11-20`, `gpt-4o-mini`, `gpt-4o-mini-2024-07-18` **GPT-5 series:** `gpt-5`, `gpt-5-2025-08-07`, `gpt-5-mini`, `gpt-5-nano`, `gpt-5.1`, `gpt-5.2`, `gpt-5.3-codex`, `gpt-5.4` **O-series (reasoning):** `o1`, `o1-2024-12-17`, `o3`, `o3-2025-04-16`, `o3-mini`, `o3-mini-2025-01-31`, `o4-mini`, `o4-mini-2025-04-16` **Image generation:** `gpt-image-2`, `gpt-image-2-vip` ### Anthropic (Claude) **Available models:** `claude-opus-4-8`, `claude-opus-4-7`, `claude-opus-4-7-thinking`, `claude-opus-4-6`, `claude-opus-4-6-thinking`, `claude-haiku-4-5-20251001`, `claude-haiku-4-5-20251001-thinking`, `claude-opus-4-5-20251101`, `claude-opus-4-5-20251101-thinking`, `claude-sonnet-4-5-20250929`, `claude-sonnet-4-5-20250929-thinking` Accessed via the native Anthropic Messages API format (`/v1/messages`), authenticated with an `x-api-key` header. ### DeepSeek **Chat:** `deepseek-chat` **Reasoning:** `deepseek-reasoner`, `deepseek-r1`, `deepseek-r1-250528` **Distilled:** `deepseek-r1-distill-llama-70b`, `deepseek-r1-distill-llama-8b` **V3 series:** `deepseek-v3`, `deepseek-v3-fast`, `deepseek-v3.1`, `deepseek-v3.1-fast`, `deepseek-v3.1-terminus`, `deepseek-v3.2`, `deepseek-v3.2-exp`, `deepseek-v3.2-fast` **Other:** `deepseek-ocr` ### Google (Gemini) **Text/Chat models:** `gemini-3-flash-preview` **Image generation models:** `gemini-2.5-flash-image`, `gemini-3-pro-image-preview` Accessed via the native Gemini `generateContent` format, authenticated with an `x-goog-api-key` header. For image generation, set `generationConfig.responseModalities` to `["image", "text"]`. ### Moonshot (Kimi) & Alibaba (Qwen) Kimi and Qwen models are available through the OpenAI-compatible `/v1/chat/completions` endpoint — just set `model` to the desired Kimi or Qwen model name. > ⚠️ **Note:** At the time of writing, the dedicated Kimi and Qwen endpoint pages in the official docs (ai-docs.tikhub.io) display placeholder content mirrored from the DeepSeek page rather than a distinct Kimi/Qwen model list. For the exact, current model identifiers, check the **Provider** filter on the [pricing page](https://ai.tikhub.io/pricing) (Alibaba / Moonshot) or query `GET /v1/models`. ### Video Generation Video generation is supported through dedicated task-based endpoints (create → poll/list → retrieve), covering: - **Sora** — e.g. `sora-2` (text/image-to-video) - **Veo** — Google's video generation models - **Kling** — text-to-video and image-to-video - **Seedance** — task-based video generation Refer to the `VideoCreateRequest`, `VeoGenerateRequest`, `KlingText2VideoRequest`, `KlingImage2VideoRequest`, and `SeedanceTaskRequest` schemas in the docs for exact request shapes. --- ## Authentication ### Standard Authentication Most endpoints use Bearer token authentication: ``` Authorization: Bearer YOUR_API_KEY ``` ### Provider-Specific Authentication Some APIs require alternative headers: **Claude Messages API** ``` x-api-key: YOUR_API_KEY ``` **Gemini API** ``` x-goog-api-key: YOUR_API_KEY ``` --- ## Base URL All API requests should be sent to: ``` https://ai.tikhub.io ``` --- ## Support For technical support, documentation inquiries, or account-related assistance, please visit: 🌐 [https://ai.tikhub.io](https://ai.tikhub.io) 📩 team@tikhub.io Our team will respond as soon as possible.