# RouterBase

- [Getting Started](https://docs.routerbase.com/index.md): Everything you need to integrate RouterBase APIs into your product.
- [Model Overview](https://docs.routerbase.com/overview.md): All available models and their model IDs.
- [Keys](https://docs.routerbase.com/essentials/keys.md): RouterBase has three kinds of keys. Pick the right one for your use case.
- [Async Tasks](https://docs.routerbase.com/essentials/async-tasks.md): Video and audio generations are asynchronous. Here's how to retrieve results.
- [Idempotency](https://docs.routerbase.com/essentials/idempotency.md): Retry media generations safely and recover tasks after a dropped connection.
- [Rate Limits](https://docs.routerbase.com/essentials/rate-limits.md): Understanding and configuring rate limits on your API keys.
- [Error Codes](https://docs.routerbase.com/essentials/errors.md): API error reference: HTTP status codes, error_code values, and retry classification.
- [Chat Completions](https://docs.routerbase.com/api-reference/chat-completions.md): OpenAI-compatible chat completions endpoint. Synchronous, with optional streaming.
- [Responses](https://docs.routerbase.com/api-reference/responses.md): OpenAI Responses API endpoint (stateless). The wire modern coding agents like the Codex CLI speak.
- [OpenAI Chat API](https://docs.routerbase.com/openai-gpt-api/quickstart.md): OpenAI chat models — GPT-5.6 (Luna/Terra/Sol), GPT-5 (5/5.4/5.4-mini/5.5/nano), GPT-4 (4o/4o-mini/4.1/4.1-nano) and the o-series (o3, o4-mini) — via RouterBase, OpenAI-compatible with per-user prompt caching.
- [Anthropic Claude API](https://docs.routerbase.com/anthropic-claude-api/quickstart.md): Claude Fable, Opus, Sonnet and Haiku via RouterBase — OpenAI-compatible chat completions with per-user prompt caching.
- [Google Gemini API](https://docs.routerbase.com/google-gemini-api/quickstart.md): Gemini 2.5 / 3.x via RouterBase — OpenAI-compatible chat completions with per-user prompt caching.
- [ZAI GLM Chat API](https://docs.routerbase.com/zai-glm-api/quickstart.md): ZAI GLM chat models — GLM-4.5 (4.5/4.5-Air/4.5V), GLM-4.6 (4.6/4.6V), GLM-4.7, and GLM-5 (5/5.1/5.2) — via RouterBase, OpenAI-compatible chat completions.
- [Moonshot Kimi Chat API](https://docs.routerbase.com/moonshot-kimi-api/quickstart.md): Moonshot AI Kimi chat models — K3, K2 Instruct, K2 0905, K2 Thinking, K2.5, and K2.6 — via RouterBase, OpenAI-compatible chat completions.
- [DeepSeek Chat API](https://docs.routerbase.com/deepseek-chat-api/quickstart.md): DeepSeek chat models — V4 (Pro/Flash), V3 (3.2/3.2-Exp/3.1/3.1-Terminus/3-0324/turbo), R1 (R1-0528/turbo/distill), and Prover-V2 — via RouterBase, OpenAI-compatible chat completions.
- [MiniMax Chat API](https://docs.routerbase.com/minimax-chat-api/quickstart.md): MiniMax chat models — M2 (M2/M2.1/M2.5/M2.7) and M1-80k — via RouterBase, OpenAI-compatible chat completions.
- [Qwen Chat API](https://docs.routerbase.com/qwen-chat-api/quickstart.md): Alibaba Qwen chat models via RouterBase, OpenAI-compatible chat completions.
- [Image Generation](https://docs.routerbase.com/api-reference/images.md): OpenAI-compatible image generation. Synchronous — the response holds until the image is ready.
- [Seedream 3.0 — Text-to-Image](https://docs.routerbase.com/bytedance-seedream-3-0-api/text-to-image.md): Generate images from text prompts with ByteDance Seedream 3.0.
- [Seedream 4.0 — Text-to-Image](https://docs.routerbase.com/bytedance-seedream-4-0-api/text-to-image.md): Generate images from text prompts with ByteDance Seedream 4.0.
- [Seedream 4.5 — Text-to-Image](https://docs.routerbase.com/bytedance-seedream-4-5-api/text-to-image.md): Generate images from text prompts with ByteDance Seedream 4.5.
- [Seedream 5.0 Lite — Text-to-Image](https://docs.routerbase.com/bytedance-seedream-5-0-lite-api/text-to-image.md): Generate images from text prompts with ByteDance Seedream 5.0 Lite.
- [Nano Banana (Gemini 2.5 Flash Image) API](https://docs.routerbase.com/google-gemini-2-5-flash-image-api/quickstart.md): Quickstart for Google Nano Banana — fast, cost-efficient text-to-image and natural-language image editing.
- [Nano Banana — Text-to-Image](https://docs.routerbase.com/google-gemini-2-5-flash-image-api/text-to-image.md): Generate images from text prompts with Google Nano Banana (Gemini 2.5 Flash Image).
- [Nano Banana — Edit](https://docs.routerbase.com/google-gemini-2-5-flash-image-api/edit.md): Edit images with natural-language prompts using Google Nano Banana (Gemini 2.5 Flash Image).
- [Gemini 3.1 Flash Image API](https://docs.routerbase.com/google-gemini-3-1-flash-image-api/quickstart.md): Quickstart for Google Gemini 3.1 Flash Image — fast text-to-image and natural-language image editing.
- [Gemini 3.1 Flash Image — Text-to-Image](https://docs.routerbase.com/google-gemini-3-1-flash-image-api/text-to-image.md): Generate images from text prompts with Google Gemini 3.1 Flash Image.
- [Gemini 3.1 Flash Image — Edit](https://docs.routerbase.com/google-gemini-3-1-flash-image-api/edit.md): Edit images with natural-language prompts using Google Gemini 3.1 Flash Image.
- [Nano Banana Pro (Gemini 3 Pro Image) API](https://docs.routerbase.com/google-gemini-3-pro-image-api/quickstart.md): Quickstart for Google Nano Banana Pro — flagship text-to-image and natural-language image editing up to 4K.
- [Nano Banana Pro — Text-to-Image](https://docs.routerbase.com/google-gemini-3-pro-image-api/text-to-image.md): Generate up-to-4K images from text prompts with Google Nano Banana Pro (Gemini 3 Pro Image).
- [Nano Banana Pro — Edit](https://docs.routerbase.com/google-gemini-3-pro-image-api/edit.md): Edit images with natural-language prompts using Google Nano Banana Pro (Gemini 3 Pro Image).
- [GPT Image 2 API](https://docs.routerbase.com/openai-gpt-image-2-api/quickstart.md): Quickstart for OpenAI GPT Image 2 — text-to-image and image editing with quality tiers.
- [GPT Image 2 — Text-to-Image](https://docs.routerbase.com/openai-gpt-image-2-api/text-to-image.md): Generate images from text prompts with OpenAI GPT Image 2.
- [GPT Image 2 — Edit](https://docs.routerbase.com/openai-gpt-image-2-api/edit.md): Edit images with natural-language prompts using OpenAI GPT Image 2.
- [Video Generation](https://docs.routerbase.com/api-reference/videos.md): Async video generation. Submit a job, then poll by id (or supply a callback_url).
- [Google Veo 3.1 API](https://docs.routerbase.com/google-veo-3-1-api/quickstart.md): Quickstart for Google Veo 3.1 — high-quality video generation from text or image prompts.
- [Google Veo 3.1 — Image to Video](https://docs.routerbase.com/google-veo-3-1-api/image-to-video.md): Animate a still image into a high-quality short video with Google Veo 3.1.
- [Google Veo 3.1 — Text to Video](https://docs.routerbase.com/google-veo-3-1-api/text-to-video.md): Generate a high-quality 8s video from a text prompt with Google Veo 3.1.
- [OpenAI Sora 2 API](https://docs.routerbase.com/openai-sora-2-api/quickstart.md): Quickstart for OpenAI Sora 2 — text-to-video and image-to-video at 720p, with synchronized audio.
- [OpenAI Sora 2 — Text to Video](https://docs.routerbase.com/openai-sora-2-api/text-to-video.md): Generate a short video with synchronized audio from a text prompt at 720p.
- [OpenAI Sora 2 — Image to Video](https://docs.routerbase.com/openai-sora-2-api/image-to-video.md): Animate a reference image with a text prompt at 720p, with synchronized audio.
- [OpenAI Sora 2 Pro API](https://docs.routerbase.com/openai-sora-2-pro-api/quickstart.md): Quickstart for OpenAI Sora 2 Pro — text-to-video and image-to-video at 720p or 1080p, with synchronized audio.
- [OpenAI Sora 2 Pro — Text to Video](https://docs.routerbase.com/openai-sora-2-pro-api/text-to-video.md): Generate a high-fidelity video with synchronized audio from a text prompt at 720p or 1080p.
- [OpenAI Sora 2 Pro — Image to Video](https://docs.routerbase.com/openai-sora-2-pro-api/image-to-video.md): Animate a reference image with a text prompt at 720p or 1080p, with synchronized audio.
- [Kling v3.0 4K — Text to Video](https://docs.routerbase.com/kuaishou-kling-v3-4k-api/text-to-video.md): Generate 4K video from a text prompt with Kuaishou Kling 3.0. Optional synthesized audio.
- [Kling v3.0 4K — Image to Video](https://docs.routerbase.com/kuaishou-kling-v3-4k-api/image-to-video.md): Animate a reference image into a 4K video with Kuaishou Kling 3.0. Optional synthesized audio.
- [Kling v3.0 Motion Control](https://docs.routerbase.com/kuaishou-kling-v3-motion-control-api/image-to-video.md): Drive a subject image's motion from either a still or a reference video using Kuaishou Kling 3.0.
- [Alibaba Wan 2.7 API](https://docs.routerbase.com/alibaba-wan-2-7-api/quickstart.md): Quickstart for Alibaba Wan 2.7 — text, image, reference, and edit video generation at 720P / 1080P.
- [Wan 2.7 — Text to Video](https://docs.routerbase.com/alibaba-wan-2-7-api/text-to-video.md): Generate a 2–15s video from a text prompt with Wan 2.7 at 720P or 1080P.
- [Wan 2.7 — Image to Video](https://docs.routerbase.com/alibaba-wan-2-7-api/image-to-video.md): Animate a still image with a text prompt using Wan 2.7.
- [Wan 2.7 — Reference to Video](https://docs.routerbase.com/alibaba-wan-2-7-api/reference-to-video.md): Generate video conditioned on up to 5 reference images and/or 3 reference videos with Wan 2.7.
- [Wan 2.7 — Video Edit](https://docs.routerbase.com/alibaba-wan-2-7-api/video-edit.md): Edit an existing video clip with a text prompt and optional reference images using Wan 2.7.
- [Doubao-Seedance 2.0 — Text to Video](https://docs.routerbase.com/bytedance-seedance-2-api/text-to-video.md): Generate a 3–12s video from a text prompt with Doubao-Seedance 2.0 at 720p or 1080p.
- [Doubao-Seedance 2.0 — Image to Video](https://docs.routerbase.com/bytedance-seedance-2-api/image-to-video.md): Animate a still image into a 3–12s video with Doubao-Seedance 2.0 at 720p or 1080p.
- [Doubao-Seedance 2.0 Fast — Text to Video](https://docs.routerbase.com/bytedance-seedance-2-fast-api/text-to-video.md): Accelerated, lower-cost Doubao-Seedance 2.0 — 720p-only text to video.
- [Doubao-Seedance 2.0 Fast — Image to Video](https://docs.routerbase.com/bytedance-seedance-2-fast-api/image-to-video.md): Accelerated, lower-cost Doubao-Seedance 2.0 — 720p-only image to video.
- [MiniMax-H3 API](https://docs.routerbase.com/minimax-h3-api/quickstart.md): Quickstart for MiniMax-H3 — text-to-video at 2K with synchronized audio, MiniMax's flagship multimodal video model.
- [MiniMax-H3 — Text to Video](https://docs.routerbase.com/minimax-h3-api/text-to-video.md): Generate a 2K video from a text prompt with MiniMax-H3, MiniMax's flagship multimodal video model.
- [MiniMax-H3 — Image to Video](https://docs.routerbase.com/minimax-h3-api/image-to-video.md): Animate a still first-frame image into a 2K video with MiniMax-H3, MiniMax's flagship multimodal video model.
- [MiniMax-H3 — Reference to Video](https://docs.routerbase.com/minimax-h3-api/reference-to-video.md): Generate a 2K video that keeps a subject/character consistent from a reference image with MiniMax-H3.
- [MiniMax-H3-Max API](https://docs.routerbase.com/minimax-h3-max-api/quickstart.md): Quickstart for MiniMax-H3-Max — long-form 5-15 second audio-inclusive video at 480P/768P from text or a first frame.
- [MiniMax-H3-Max — Text to Video](https://docs.routerbase.com/minimax-h3-max-api/text-to-video.md): Generate a 5-15 second audio-inclusive video from a text prompt with MiniMax-H3-Max.
- [MiniMax-H3-Max — Image to Video](https://docs.routerbase.com/minimax-h3-max-api/image-to-video.md): Animate a first frame into a 5-15 second audio-inclusive video with MiniMax-H3-Max.
- [Vidu Q1 — Text to Video](https://docs.routerbase.com/vidu-q1-api/text-to-video.md): Vidu Q1 t2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q1 — Image to Video](https://docs.routerbase.com/vidu-q1-api/image-to-video.md): Vidu Q1 i2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q1 — Reference to Video](https://docs.routerbase.com/vidu-q1-api/reference-to-video.md): Vidu Q1 r2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q2 — Text to Video](https://docs.routerbase.com/vidu-q2-api/text-to-video.md): Vidu Q2 t2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q2 — Reference to Video](https://docs.routerbase.com/vidu-q2-api/reference-to-video.md): Vidu Q2 r2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q2 Pro — Image to Video](https://docs.routerbase.com/vidu-q2-pro-api/image-to-video.md): Vidu Q2 Pro i2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q2 Pro Fast — Reference to Video](https://docs.routerbase.com/vidu-q2-pro-fast-api/reference-to-video.md): Vidu Q2 Pro Fast r2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q2 Turbo — Image to Video](https://docs.routerbase.com/vidu-q2-turbo-api/image-to-video.md): Vidu Q2 Turbo i2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q3 Pro — Text to Video](https://docs.routerbase.com/vidu-q3-pro-api/text-to-video.md): Vidu Q3 Pro t2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q3 Pro — Image to Video](https://docs.routerbase.com/vidu-q3-pro-api/image-to-video.md): Vidu Q3 Pro i2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q3 Pro — Start-End Frame to Video](https://docs.routerbase.com/vidu-q3-pro-startend-api/image-to-video.md): Vidu Q3 Pro f2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q3 Turbo — Text to Video](https://docs.routerbase.com/vidu-q3-turbo-api/text-to-video.md): Vidu Q3 Turbo t2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q3 Turbo — Image to Video](https://docs.routerbase.com/vidu-q3-turbo-api/image-to-video.md): Vidu Q3 Turbo i2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Vidu Q3 Turbo — Start-End Frame to Video](https://docs.routerbase.com/vidu-q3-turbo-startend-api/image-to-video.md): Vidu Q3 Turbo f2v generation on RouterBase via one OpenAI-compatible endpoint.
- [Embeddings](https://docs.routerbase.com/api-reference/embeddings.md): OpenAI-compatible embeddings endpoint. Returns dense vectors for search, clustering, and RAG.
- [OpenAI Text Embeddings API](https://docs.routerbase.com/openai-text-embedding-api/quickstart.md): OpenAI text-embedding-3-large via RouterBase — OpenAI-compatible embeddings for search, clustering, and RAG.
- [Models API](https://docs.routerbase.com/api-reference/models.md): List and query available models and their pricing.
- [API Keys](https://docs.routerbase.com/api-reference/api-keys.md): Create, list, update, and revoke General API Keys — the keys your end users hold to call completion endpoints.
- [BYOK](https://docs.routerbase.com/api-reference/byok.md): Bring Your Own Key — register upstream provider credentials so RouterBase uses your account to call OpenAI / Anthropic / Google on your behalf.
- [Billing](https://docs.routerbase.com/api-reference/billing.md): Check your USD balance, top up via Stripe, and view transaction history.