# APIMart

## Docs

- [APIMart — OpenAI-Compatible API Gateway (GPT-5, Claude, Gemini)](https://docs.apimart.ai/en/index.md): OpenAI-compatible API for GPT-5, Claude, Gemini. Multi-provider routing, transparent pricing, low latency. Enterprise SLA, SDK support, pay-as-you-go.
- [General Chat API (Default Streaming)](https://docs.apimart.ai/en/api-reference/texts/general/chat-completions.md): - Unified chat API interface supporting all text generation models - Select different AI models via the model parameter - Compatible with OpenAI Chat Completions API format
- [General Chat API (Default Non-Streaming)](https://docs.apimart.ai/en/api-reference/texts/general/chat-completions-nostream.md): - Unified chat API interface supporting all text generation models - Select different AI models via the model parameter - Compatible with OpenAI Chat Completions API format - Non-streaming output, returns complete response at once
- [Claude Messages API](https://docs.apimart.ai/en/api-reference/texts/general/claude-messages.md): - Fully compatible with the native Anthropic Claude Messages protocol (`POST /v1/messages`) - Supports multi-turn conversations, streaming SSE, tool use, and extended thinking - Supports multimodal content including text and images - Responses are upstream passthrough with no `{code, data}` wrapper
- [Gemini Native Format](https://docs.apimart.ai/en/api-reference/texts/gemini/quickstart.md): - Call Gemini models using Google Native API format - Synchronous processing mode with real-time response - Minimal parameters for quick start
- [OpenAI Multimodal Responses API](https://docs.apimart.ai/en/api-reference/texts/openai/responses.md): - Fully compatible with OpenAI Responses API format - Supports multimodal input with text and images - Supports tool extensions: web search, file search, function calling, remote MCP
- [Gemini Context Caching Guide](https://docs.apimart.ai/en/api-reference/texts/gemini/context-cache.md): Create and reuse Gemini context caches (Context Cache) through the OpenAI-compatible Chat Completions API or the native Gemini API. Use cache_control to cache stable prefixes and reduce token costs for repeated long content.
- [Claude Context Caching Guide](https://docs.apimart.ai/en/api-reference/texts/general/claude-context-cache.md): Cache reusable prompt prefixes through the Claude Messages API or the OpenAI-compatible Chat Completions API to reduce token costs from repeatedly processing long content.
- [Nano banana2 Image Generation](https://docs.apimart.ai/en/api-reference/images/gemini-3.1-flash/generation.md): - Supports text-to-image and image-to-image, up to 4K resolution output - Up to 14 reference images for style/character consistency - Supports extreme aspect ratios (1:4, 4:1, 1:8, 8:1) - Integrated Google Search enhancement for more realistic image generation
- [Nano Banana Lite Image Generation](https://docs.apimart.ai/en/api-reference/images/gemini-3.1-flash/generation-lite.md): - The fastest and cheapest image model in the Gemini 3.1 family, designed for large-scale, low-cost image generation - Supports 1K resolution only (passing 2K/4K/0.5K is automatically downgraded to 1K without an error) - Supports text-to-image and image-to-image, up to 14 reference images - Billed b…
- [Nano banana Pro Image Generation](https://docs.apimart.ai/en/api-reference/images/gemini-3-pro/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - High-quality image generation, professional-grade image creation - Generated image links are valid for 24 hours, please save them promptly
- [Nano banana Image Generation](https://docs.apimart.ai/en/api-reference/images/gemini-2.5-flash/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Fast generation speed, optimized for quick image creation - Generated image links are valid for 24 hours, please save them promptly
- [Imagen 4.0 Apimart Image Generation](https://docs.apimart.ai/en/api-reference/images/imagen-4.0-apimart/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Based on Google Imagen 4 model, high-quality image generation - Only supports text-to-image, does not support image-to-image / reference images - Supported ratios: 16:9 (landscape), 9:16 (portrait) - Generated image links are v…
- [GPT-Image(1/1.5) Image Generation](https://docs.apimart.ai/en/api-reference/images/gpt-image-1/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports text-to-image, image-to-image, and inpainting generation modes - Supports transparent backgrounds, multiple output formats, and quality tiers - Generate up to 4 images per request, with up to 15 reference images
- [GPT-Image-2 Image Generation](https://docs.apimart.ai/en/api-reference/images/gpt-image-2/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - OpenAI Images compatible protocol, supports text-to-image / image-to-image - 15 image aspect ratios supported via the `size` field - Output pixel tier controlled via `resolution` (`1k` / `2k` / `4k`)
- [GPT-Image-2 Official Channel Image Generation](https://docs.apimart.ai/en/api-reference/images/gpt-image-2/official.md): - OpenAI official `gpt-image-2` model, based on `/v1/images/generations` compatible protocol - Asynchronous processing, returns `task_id` for subsequent queries - Text-to-image / image-to-image / inpainting (mask) — all-in-one - New `resolution` tier field — 1K / 2K / 4K selection - 15 aspect ratios…
- [Seedream-4.0 Image Generation](https://docs.apimart.ai/en/api-reference/images/seedream-4/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports multiple generation modes including text-to-image, image-to-image, and image editing - Generated image links are valid for 24 hours, please save them promptly
- [Seedream-4.5 Image Generation](https://docs.apimart.ai/en/api-reference/images/seedream-4.5/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports multiple generation modes including text-to-image, image-to-image, and image editing - Generated image links are valid for 24 hours, please save them promptly
- [Seedream-5.0-Lite Image Generation](https://docs.apimart.ai/en/api-reference/images/seedream-5-lite/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports multiple generation modes including text-to-image, image-to-image, and sequential image generation - Supports 2K / 3K resolution with PNG / JPEG output formats - Generated image links are valid for 72 hours, please sav…
- [Seedream-5.0-Pro Image Generation](https://docs.apimart.ai/en/api-reference/images/seedream-5-0-pro/generation.md): - Asynchronous processing mode, returns a task ID for subsequent queries - Supports text-to-image, single-image-to-image, and multi-reference image-to-image (up to 10 reference images) - Supports 1K / 1.5K / 2K resolution tiers, or exact pixels via `size` - Single-image model: one image per request;…
- [FLUX Kontext Image Generation and Editing](https://docs.apimart.ai/en/api-reference/images/flux-kontext/generation.md): Submit asynchronous FLUX Kontext image generation or image editing tasks. The API returns a task ID; poll the task endpoint for the generated image.
- [FLUX.2 Image Generation](https://docs.apimart.ai/en/api-reference/images/flux-2/generation.md): Submit asynchronous FLUX.2 text-to-image or reference-image generation tasks. The API returns a task ID; poll the task endpoint for the generated image.
- [Qwen Image 2.0 Image Generation](https://docs.apimart.ai/en/api-reference/images/qwen-image/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports text-to-image, image-to-image and other generation modes - Supports 1K/2K resolution tiers, generate up to 6 images
- [Qwen Image 3.0 Image Generation](https://docs.apimart.ai/en/api-reference/images/qwen-image-3.0/generation.md): - Asynchronous processing mode, returns a task ID for subsequent queries - Supports text-to-image and image-to-image (1-3 reference images for editing) - Supports 1K / 2K resolution, up to 6 images per request - Standard and Pro variants (Pro is stronger for dense layout and text)
- [Z-Image-Turbo Image Generation](https://docs.apimart.ai/en/api-reference/images/z-image-turbo/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Lightweight and fast image generation, supports Chinese and English - Supports 1K/2K resolution tiers, supports smart prompt rewriting
- [Grok Imagine 1.5 Image Generation](https://docs.apimart.ai/en/api-reference/images/grok-imagine/generation.md): - Async processing mode, returns task ID for subsequent queries - Supports text-to-image and reference-based image editing - Generated image links are valid for 24 hours; please save them promptly
- [wan2.7 Image Generation & Editing](https://docs.apimart.ai/en/api-reference/images/wan2.7-image/generation.md): - Wan2.7 image series: supports text-to-image, image editing, interactive editing, sequential generation, and multi-image reference - Asynchronous processing mode — submit a task and poll for results using the returned task_id - Supports 1K / 2K / 4K resolution; wan2.7-image-pro supports up to 4K fo…
- [Midjourney API overview](https://docs.apimart.ai/en/api-reference/images/midjourney/generation.md): - Overview of Midjourney text-to-image (Imagine) / image-guided / follow-up actions / image-to-video endpoints - Async task mode: receive task_id after submission, poll for the result - New routes auto-inject model=midjourney and support native MJ args, structured body fields, and metadata
- [Imagine (text-to-image)](https://docs.apimart.ai/en/api-reference/images/midjourney/imagine.md): Midjourney text-to-image / image-guided generation. The default entry /v1/midjourney/generations and the explicit /imagine entry behave the same
- [Blend (multi-image blend)](https://docs.apimart.ai/en/api-reference/images/midjourney/blend.md): Blend 2–4 images into a new image (the classic MJ blend); image-only, no prompt supported
- [Describe (image to text)](https://docs.apimart.ai/en/api-reference/images/midjourney/describe.md): Reverse a prompt from an image; responds synchronously (1–3s), result in the prompt / description fields
- [Edits (image edit)](https://docs.apimart.ai/en/api-reference/images/midjourney/edits.md): Rewrite the whole image from an existing image + prompt. Good for background replacement, style transfer, content changes
- [Upscale](https://docs.apimart.ai/en/api-reference/images/midjourney/upscale.md): Pick one of U1–U4 from an Imagine grid to produce a single image; composed locally and usually returns instantly
- [Variation](https://docs.apimart.ai/en/api-reference/images/midjourney/variation.md): Subtle variation (varySubtle, equivalent to V1–V4) on one tile of an Imagine grid
- [High Variation](https://docs.apimart.ai/en/api-reference/images/midjourney/high-variation.md): Strong variation (varyStrong, Vary (Strong)) on a single upscaled image
- [Low Variation](https://docs.apimart.ai/en/api-reference/images/midjourney/low-variation.md): Subtle variation (varySubtle, same behavior as Variation, only the billing key differs) on a single upscaled image
- [Reroll](https://docs.apimart.ai/en/api-reference/images/midjourney/reroll.md): Regenerate 4 images from the source task's prompt (🔄). The whole grid is re-rolled, no index needed
- [Zoom](https://docs.apimart.ai/en/api-reference/images/midjourney/zoom.md): Zoom Out (outpaint) on a single upscaled image: the original is kept and more background is filled outward (Outpaint / CustomZoom)
- [Pan](https://docs.apimart.ai/en/api-reference/images/midjourney/pan.md): Pan out in a direction on a single upscaled image; chain pans to stitch a panorama (v6 / v6.1 / v7 / v8.1 / v8.2 / niji 6 only)
- [Inpaint](https://docs.apimart.ai/en/api-reference/images/midjourney/inpaint.md): Region inpaint entry (Vary (Region)); after submission the task enters MODAL, then call modal with mask + prompt
- [Modal (submit parameters)](https://docs.apimart.ai/en/api-reference/images/midjourney/modal.md): Supply mask + prompt to complete a MODAL-state inpaint task
- [Video (image-to-video)](https://docs.apimart.ai/en/api-reference/images/midjourney/video.md): Midjourney image-to-video (i2v), fixed FAST, no t2v support, ~5 second duration
- [Remix (reshape, v8.1 / v8.2 only)](https://docs.apimart.ai/en/api-reference/images/midjourney/remix.md): The v8 panel's reshape: regenerates the parent image and can change the prompt, in strong / subtle strengths
- [Get task](https://docs.apimart.ai/en/api-reference/images/midjourney/query.md): Query Midjourney task status and results. Unified task API /v1/tasks/{task_id} and MJ-style API /v1/midjourney/{task_id}
- [Best Practices](https://docs.apimart.ai/en/api-reference/images/midjourney/best-practices.md): Polling patterns, prompt design, image guidance, error-retry strategies, concurrency, and troubleshooting tips for Midjourney integration
- [End-to-End Workflow Examples](https://docs.apimart.ai/en/api-reference/images/midjourney/workflow.md): End-to-end curl walkthroughs for imagine → upscale → inpaint → video, with bash / Python / TS client wrappers
- [doubao-seedance-1-0-pro Video Generation](https://docs.apimart.ai/en/api-reference/videos/doubao/generation.md): - Async processing mode, returns task ID for subsequent queries - Supports text-to-video, image-to-video (first frame/last frame) - Supports landscape, portrait, and square aspect ratios
- [doubao-seedance-1-5-pro Video Generation](https://docs.apimart.ai/en/api-reference/videos/doubao-seedance-1-5-pro/generation.md): - Async processing mode, returns task ID for subsequent queries - Supports text-to-video, image-to-video (first frame/last frame) - Supports audio generation - Supports landscape, portrait, and square aspect ratios
- [doubao-seedance-2.0 Video Generation](https://docs.apimart.ai/en/api-reference/videos/doubao-seedance-2-0/generation.md): - Async processing mode, returns task ID for subsequent queries - Supports text-to-video, image-to-video (first frame/last frame) - Supports reference video, reference audio, audio-enabled video - Supports landscape, portrait, square, ultra-wide, and adaptive aspect ratios
- [Virtual Avatar Assets](https://docs.apimart.ai/en/api-reference/videos/doubao-seedance-2-0/private-avatar.md): - Private-domain virtual avatar asset submission API - Supports batch submission, up to 20 assets per request - Automatically creates or reuses asset groups; returns a task ID for status polling - Approved assets can be used directly in Seedance 2.0 video generation
- [doubao-seedance-2.5 Video Generation](https://docs.apimart.ai/en/api-reference/videos/doubao-seedance-2-5/generation.md): - Async API; returns task_id for polling - Text-to-video / multimodal reference / edit / extend / first–last frame - Up to 30s per job; up to 30 images + 10 videos + 10 audios as references - Resolution 480p / 720p only; mp4 or mov output
- [Sora2 Video Generation](https://docs.apimart.ai/en/api-reference/videos/sora-2/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports multiple generation modes including text-to-video and image-to-video - Supports both `sora-2` and `sora-2-pro` models - Generated video links are valid for 24 hours, please save them promptly
- [VEO3 Video Generation](https://docs.apimart.ai/en/api-reference/videos/veo3/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports multiple generation modes including text-to-video and image-to-video - Supports **4K** resolution output - Generated video links are valid for 24 hours, please save them promptly
- [VEO3 Video Remix](https://docs.apimart.ai/en/api-reference/videos/veo3/remix.md): - Extend generated videos by continuing from 8 seconds to 15 seconds - Asynchronous processing mode, returns task ID for subsequent queries - Generated video links are valid for 24 hours, please save them promptly
- [VEO3 Official Video Generation](https://docs.apimart.ai/en/api-reference/videos/veo3/generation-official.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports text-to-video and image-to-video (first frame / first & last frame control) - Supports 720P and 1080P resolution - Supports 4/6/8 second video duration - Supports audio track generation - Supports person generation pol…
- [MiniMax-Hailuo-02 Video Generation](https://docs.apimart.ai/en/api-reference/videos/minimax-hailuo/generation.md): - Asynchronous processing mode, returns task ID for subsequent queries - Supports text-to-video, image-to-video (first frame/last frame) - Supports 5 and 10 second durations, multiple resolutions available - Supports automatic prompt optimization and watermark control
- [MiniMax-Hailuo-2.3 Video Generation](https://docs.apimart.ai/en/api-reference/videos/minimax-hailuo-2.3/generation.md): - Async processing mode, returns task ID for subsequent queries - Supports text-to-video, image-to-video (first frame image) - Supports 6s and 10s duration, 768p/1080p resolution - Supports 15 camera movement commands, prompt auto-optimization, and watermark control
- [MiniMax-H3 Video Generation](https://docs.apimart.ai/en/api-reference/videos/minimax-h3/generation.md): - Async processing mode, returns a task ID for subsequent queries - Supports text-to-video, image-to-video (first / last / first+last frame), and multimodal reference-to-video (reference images + videos + audio) - Supports 2K / 768P resolution, duration 4 ~ 15 seconds, with audio track - Shares the…
- [MiniMax-H3 Context-IR Prompt Enhancement](https://docs.apimart.ai/en/api-reference/videos/minimax-h3/context-ir.md): - Multimodal context understanding that produces an enhanced structured prompt (text only, no video) - Shares the same media fields and mutual-exclusion rules as H3 video generation - Token-based billing; typically completes in 20~40 seconds - Use alone, or as step 1 of the 768P preview → 2K regener…
- [MiniMax-H3 Regeneration](https://docs.apimart.ai/en/api-reference/videos/minimax-h3/regeneration.md): - Regenerate MiniMax-H3 768P video into 2K - Prefer source_task_id only; the platform auto-fills prompt, media, and source video - Not generic upscaling: source must be an H3 768P output - Per-second billing ($0.045/s output + same rate for reference videos)
- [FLUX 3 Video Generation](https://docs.apimart.ai/en/api-reference/videos/flux-3-video/generation.md): - Asynchronous processing mode, returns a task ID for subsequent queries - Unified entry: text-to-video / image-to-video / video continuation / draft two-step - Output H.264 + AAC with synced audio, duration 5~20 seconds - Resolution hd / fhd, seven aspect ratios
- [SkyReels V4 Video Generation](https://docs.apimart.ai/en/api-reference/videos/skyreels-v4/generation.md): - Two model tiers: Fast (speed-optimized) and Std (quality-optimized) - Three modes auto-routed by request fields: Text-to-Video (T2V), Image-to-Video (I2V), Multimodal Reference (Omni) - 480p / 720p / 1080p resolution, 3 ~ 15 seconds duration - Advanced features: first/end/key frame, reference imag…
- [HappyHorse 1.0 Video Generation](https://docs.apimart.ai/en/api-reference/videos/happyhorse-1.0/generation.md): - Alibaba Cloud Bailian HappyHorse 1.0 video generation model (unified entry, single-model auto-routing) - Auto-routes by parameters: T2V (prompt only) / I2V (first_frame_image) / R2V (image_urls) / EDIT (video_url) - Supports 720P/1080P resolutions and any integer duration from 3 to 15 seconds - Bi…
- [HappyHorse 1.1 Video Generation](https://docs.apimart.ai/en/api-reference/videos/happyhorse-1.1/generation.md): - Alibaba Cloud Bailian HappyHorse 1.1 video generation model (unified entry, single-model auto-routing) - Auto-routes by parameters: T2V (prompt only) / I2V (first_frame_image) / R2V (image_urls) - Supports 720P/1080P resolutions and any integer duration from 3 to 15 seconds - Billed by resolution…
- [Wan2.7 Video Generation](https://docs.apimart.ai/en/api-reference/videos/wan2.7/generation.md): - Alibaba Cloud Wanxiang 2.7 video generation model (unified entry) - Automatically routed based on parameters: Text-to-Video / Image-to-Video (first frame, first-last frame, video continuation) - Supports 720P/1080P resolution, 2-15 seconds duration - Supports custom audio (background music in text…
- [Wan2.7-R2V Reference-to-Video](https://docs.apimart.ai/en/api-reference/videos/wan2.7-r2v/generation.md): - Alibaba Cloud Wanxiang 2.7 reference-to-video model - Generate a new video with consistent style, characters, and scenes based on one or more reference images/videos - Supports character consistency, style transfer, and multi-asset combination - Supports reference voice (reference_voice) to contro…
- [Wan2.7-VideoEdit Video Editing](https://docs.apimart.ai/en/api-reference/videos/wan2.7-videoedit/generation.md): - Alibaba Cloud Wanxiang 2.7 video editing model - AI editing based on existing videos: style transfer, content replacement, element addition - Optionally accepts reference images to specify target style or appearance - Supports keeping the original video duration or customizing the output duration
- [Wan2.6 Video Generation](https://docs.apimart.ai/en/api-reference/videos/wan2.6/generation.md): - Alibaba Cloud Wanxiang video generation model - Supports Text-to-Video and Image-to-Video - Supports 720p/1080p resolution, 5/10/15 seconds duration - Supports automatic prompt extension and audio generation
- [wan2.6-i2v-flash Image-to-Video](https://docs.apimart.ai/en/api-reference/videos/wan2.6/i2v-flash-generation.md): - Wanxiang 2.6 fast image-to-video model - Generates smooth video from first-frame image and text prompts - Supports audio/silent toggle, multi-shot narration, custom audio - Supports 720p/1080p resolution, 2-15 seconds duration - Supports video effect templates
- [wan2.5-preview Video Generation](https://docs.apimart.ai/en/api-reference/videos/wan2.5/generation.md): - Wanxiang 2.5 preview video generation model - Supports Text-to-Video and Image-to-Video - Supports 480p/720p/1080p resolution, 5 or 10 seconds duration - Supports auto prompt extension, auto audio and custom audio
- [Kling 2.6 Video Generation](https://docs.apimart.ai/en/api-reference/videos/kling-v2-6/generation.md): - Async processing mode, returns task ID for subsequent queries - Supports text-to-video, image-to-video (first frame/first-last frame control) - Supports standard mode (720P) and professional mode (1080P) - Professional mode supports automatic audio generation and voice selection
- [Kling v2.6 Motion Control Video Generation](https://docs.apimart.ai/en/api-reference/videos/kling-v2-6/kling-v2-6-motion-control-generation.md): - Kling motion control model (reference image + reference video) - Called via unified endpoint `/v1/videos/generations` - Supports image / video character orientation, max duration 10s / 30s respectively - Asynchronous task — returns task_id on submission
- [Kling v3 Video Generation](https://docs.apimart.ai/en/api-reference/videos/kling-v3/generation.md): - Async processing mode, returns task ID for subsequent queries - Supports text-to-video, image-to-video (first frame/first-last frame control) - Supports standard mode (720P), professional mode (1080P), and 4K mode - Supports video durations from 3 to 15 seconds - Supports generating videos with au…
- [Kling 3.0 Turbo Video Generation](https://docs.apimart.ai/en/api-reference/videos/kling-3.0-turbo/generation.md): - Asynchronous processing mode, returns a task ID for subsequent queries - Supports text-to-video and image-to-video (first-frame control) - Supports two resolution tiers: 720P / 1080P - Supports video durations of 3-15 seconds - Supports multi-shot storyboards (expressed via a fixed-format prompt)
- [Kling v3 Omni Video Generation](https://docs.apimart.ai/en/api-reference/videos/kling-v3-omni/generation.md): - Async processing mode, returns task ID for subsequent queries - Unified text-to-video/image-to-video interface with image reference syntax - Supports standard mode (720P), professional mode (1080P), and 4K mode - Reference images in prompts using image_N syntax - Supports generating videos with au…
- [Kling Video O1 Video Generation](https://docs.apimart.ai/en/api-reference/videos/kling-video-o1/generation.md): - Reasoning-enhanced model for highest quality video generation - Async processing mode, returns task ID for subsequent queries - Unified text-to-video/image-to-video interface with image reference syntax - Supports standard mode (720P) and professional mode (1080P) - Reference images in prompts usi…
- [Vidu Q3(pro/turbo) Video Generation](https://docs.apimart.ai/en/api-reference/videos/vidu-q3-pro/generation.md): - Async processing mode, returns task ID for subsequent queries - Supports text-to-video, image-to-video, first-last frame video generation - Supports 540p / 720p / 1080p resolution - Duration range 1-16 seconds, audio enabled by default
- [Vidu Q3(mix/standard) Reference-to-Video](https://docs.apimart.ai/en/api-reference/videos/vidu-q3/generation.md): - Async processing mode, returns task ID for subsequent queries - Upload 1-7 reference images + text prompt to generate short videos containing reference subjects - Supports 540p / 720p / 1080p resolution - Duration range 1-16 seconds, suitable for character consistency and style continuation
- [Grok Imagine 1.5 Video Generation](https://docs.apimart.ai/en/api-reference/videos/grok-imagine/generation.md): - Async processing mode, returns task ID for subsequent queries - High-quality AI video generation, supports text-to-video and image-to-video - Flexible aspect ratio and quality options for different creative needs
- [Pixverse v6 Video Generation](https://docs.apimart.ai/en/api-reference/videos/pixverse-v6/generation.md): - Pixverse v6 unified video generation model - Supports text-to-video, image-to-video, first/last frame transition, multi-reference fusion, and video extension - Supports 360p/540p/720p/1080p resolutions with 1-15 seconds duration - Asynchronous task API; query the result by task ID after submission
- [Omni-Flash-Ext Video Generation](https://docs.apimart.ai/en/api-reference/videos/omni-flash-ext/generation.md): - Omni-Flash-Ext unified video generation model - Supports Text-to-Video, single-image Image-to-Video, reference video, and 3-reference-image fusion - Supports 720p/1080p/4k resolution and 4/6/8/10 second duration - Asynchronous task API. Submit a task first, then query the result by task ID
- [Gemini Omni Flash Video Generation](https://docs.apimart.ai/en/api-reference/videos/gemini-omni-flash-preview/generation.md): - Google's official Gemini Omni Flash all-in-one multimodal video generation model - Supports Text-to-Video, Image-to-Video, and Video-to-Video (editing), with mixed text + image + video input - Outputs 720p / 24fps, 3-10 seconds, with audio; supports conversational multi-turn editing - Asynchronous…
- [Whisper-1 Audio Transcription](https://docs.apimart.ai/en/api-reference/audios/whisper-1.md): - Supports speech recognition in 99 languages - Multiple output formats: json, text, srt, vtt, etc. - Maximum file size 25 MB
- [TTS Text-to-Speech](https://docs.apimart.ai/en/api-reference/audios/tts.md): - Support multiple voice models and voice selections - Output high-quality audio formats: wav, opus, aac, flac, pcm - Maximum input text of 4096 characters
- [Generate Music](https://docs.apimart.ai/en/api-reference/audios/flow-music/music.md): Flow Music text-to-music generation. Supports style prompts / lyrics / BPM / duration control, generating one track per request
- [Generate Lyrics](https://docs.apimart.ai/en/api-reference/audios/flow-music/lyrics.md): Flow Music generates lyrics from a prompt. The result can be filled into the lyrics field of the Generate Music endpoint
- [Extend Music](https://docs.apimart.ai/en/api-reference/audios/flow-music/extend.md): Flow Music continues a previously generated audio clip. Extends the music from a specified time point according to an editing instruction
- [Replace Section](https://docs.apimart.ai/en/api-reference/audios/flow-music/replace.md): Flow Music replaces a section of previously generated audio. Regenerates the segment between start_s and end_s according to an editing instruction
- [Cover Rearrangement](https://docs.apimart.ai/en/api-reference/audios/flow-music/cover.md): Flow Music rearranges an entire song into a new style (cover / restyle), with strength controlling the editing intensity
- [Stem Separation](https://docs.apimart.ai/en/api-reference/audios/flow-music/stems.md): Flow Music separates vocal / instrumental tracks, delivered as a zip stem package
- [Upload Audio](https://docs.apimart.ai/en/api-reference/audios/flow-music/upload-audio.md): Import external audio into Flow Music in exchange for a clip_id for subsequent extend / replace / cover / stem operations
- [Download Audio](https://docs.apimart.ai/en/api-reference/audios/flow-music/download-audio.md): Flow Music exports a clip as an audio file in the specified format (wav / mp3)
- [Music Video Rendering](https://docs.apimart.ai/en/api-reference/audios/flow-music/video-clip.md): Flow Music renders music into an mp4 video from a template, supporting three presets: simple / modern / player
- [Query Task](https://docs.apimart.ai/en/api-reference/audios/flow-music/query.md): Query the execution status, progress, and generated results of Flow Music asynchronous tasks
- [Suno Common Conventions & Task Query](https://docs.apimart.ai/en/api-reference/audios/suno/overview.md): - Common notes for the Suno music API: authentication, async task lifecycle, model / version, source track references - Task query: GET /v1/music/tasks/:task_id, poll until completed / failed
- [Generate music](https://docs.apimart.ai/en/api-reference/audios/suno/generation.md): - Generate a song from a prompt: `custom=false` is inspiration mode (`prompt` used as an inspiration prompt), `=true` is custom mode (`prompt` used as lyrics). - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Generate lyrics](https://docs.apimart.ai/en/api-reference/audios/suno/lyrics.md): - Generate lyrics text based on a theme. - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Inspiration generation (inspo)](https://docs.apimart.ai/en/api-reference/audios/suno/inspo.md): - Generate a new song using 1–4 pieces of public audio as inspiration references (pass the audio URLs directly, not via task_id). - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Sound effect generation](https://docs.apimart.ai/en/api-reference/audios/suno/sounds.md): - Generate a sound effect from a description. - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Tag enhancement](https://docs.apimart.ai/en/api-reference/audios/suno/upsample-tags.md): - Optimize / expand style tags to improve prompt quality. - Synchronous endpoint: the result is ready immediately after submitting — no polling wait is required (you may still query `GET /v1/music/tasks/:task_id`).
- [Upload audio](https://docs.apimart.ai/en/api-reference/audios/suno/upload.md): - Import a piece of public audio to obtain a track that can be referenced later (for cover / continuation, etc.); once complete, this job's `task_id` can serve as the source (`audio_index=1`). - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Continuation and extension](https://docs.apimart.ai/en/api-reference/audios/suno/extend.md): - Continue and extend an existing song from a given point in time. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Style cover](https://docs.apimart.ai/en/api-reference/audios/suno/cover-song.md): - Cover an existing song in a different style. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Mastering](https://docs.apimart.ai/en/api-reference/audios/suno/remaster.md): - Master an already generated song to improve audio quality, clarity, and overall texture. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Stem extraction](https://docs.apimart.ai/en/api-reference/audios/suno/stems.md): - Separate a specified track (e.g. vocals) from a song. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Full stem separation](https://docs.apimart.ai/en/api-reference/audios/suno/stems-all.md): - Full multi-track separation. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Add vocals](https://docs.apimart.ai/en/api-reference/audios/suno/add-vocals.md): - Layer vocals onto an existing (accompaniment / track). - Source track must be your own track uploaded via `uploadTask`: pass that upload task's `task_id` + `audio_index`; using a generated track as the source will fail - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_i…
- [Add accompaniment](https://docs.apimart.ai/en/api-reference/audios/suno/add-instrumental.md): - Layer accompaniment onto an existing (vocals / track). - Source track must be your own track uploaded via `uploadTask`: pass that upload task's `task_id` + `audio_index`; using a generated track as the source will fail - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_i…
- [Add stem (add stem)](https://docs.apimart.ai/en/api-reference/audios/suno/add-stem.md): - Layer a stem onto an existing track. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Extract Vox](https://docs.apimart.ai/en/api-reference/audios/suno/vox.md): - Extract vocal segments from a song, producing vox reusable by Persona. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Create voice](https://docs.apimart.ai/en/api-reference/audios/suno/create-voice.md): - Create a reusable voice from a track. - Referencing the source track: this endpoint does NOT use `task_id` + `audio_index` — provide a publicly accessible `audio_url` directly - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Persona](https://docs.apimart.ai/en/api-reference/audios/suno/persona.md): - Create a singer Persona from a song; can be bound to the result of "Extract Vox". - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Section replacement](https://docs.apimart.ai/en/api-reference/audios/suno/replace-music.md): - Replace a section of a song (infill). - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Remove segment](https://docs.apimart.ai/en/api-reference/audios/suno/remove-section.md): - Remove a time range from a song. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Crop audio](https://docs.apimart.ai/en/api-reference/audios/suno/crop.md): - Crop to keep the specified range. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Fade in](https://docs.apimart.ai/en/api-reference/audios/suno/fade-in.md): - Fade in at the beginning. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Fade out](https://docs.apimart.ai/en/api-reference/audios/suno/fade-out.md): - Fade out at the end. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Adjust speed](https://docs.apimart.ai/en/api-reference/audios/suno/adjust-speed.md): - Change speed (without changing pitch). - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Full song synthesis / concatenation](https://docs.apimart.ai/en/api-reference/audios/suno/concat.md): - Synthesize segments into a complete song. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Generate mashup (mashup)](https://docs.apimart.ai/en/api-reference/audios/suno/mashup.md): - Remix a song into a new creation. - Referencing source tracks: requires **exactly 2**, specified via `task_ids` (an array of length 2) + optional `audio_indexes` - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Sample to song (sample)](https://docs.apimart.ai/en/api-reference/audios/suno/sample.md): - Generate a song based on a sample. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Generate MIDI](https://docs.apimart.ai/en/api-reference/audios/suno/midi.md): - Generate MIDI from a song. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Lyrics timeline](https://docs.apimart.ai/en/api-reference/audios/suno/aligned-lyrics.md): - Generate a line-by-line aligned lyrics timeline. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [BPM analysis](https://docs.apimart.ai/en/api-reference/audios/suno/bpm.md): - Analyze a song's BPM. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Generate music video (MV)](https://docs.apimart.ai/en/api-reference/audios/suno/generate-mp4.md): - Generate an MV video for the song. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [Export WAV](https://docs.apimart.ai/en/api-reference/audios/suno/wav.md): - Export a song as WAV. - Reference the source track: specified by `task_id` + `audio_index`, no need to record any extra id - Async task: submitting returns a `task_id`; poll `GET /v1/music/tasks/:task_id` for the result
- [omni-moderation-latest Content Moderation](https://docs.apimart.ai/en/api-reference/moderations/omni-moderation-latest/generation.md): - Supports text, image, and mixed text+image content moderation - Compatible with single text, text array, and content block array inputs - Images support public URLs and base64 Data URIs
- [Upload Image](https://docs.apimart.ai/en/api-reference/uploads/images.md): Upload an image to get a URL for use with image/video generation APIs
- [Get Task Status](https://docs.apimart.ai/en/api-reference/tasks/status.md): - Query the execution status and result of an asynchronous task - Real-time status updates and progress tracking - Retrieve generation results when tasks are completed - Multi-language support (zh/en/ko/ja)
- [Task Completion Callback (Webhook)](https://docs.apimart.ai/en/api-reference/tasks/webhook.md): Include a callback URL when submitting an async generation task, and we'll POST the result to you once the task finishes (success or failure) — no polling required.
- [Query Token Balance](https://docs.apimart.ai/en/api-reference/account/token-balance.md): - Query remaining and used credits of the current API Key - Monitor single token usage - CORS support for cross-origin requests - Real-time balance monitoring
- [Query User Balance](https://docs.apimart.ai/en/api-reference/account/user-balance.md): - Query overall remaining and used credits of user account - Get user-level balance information - CORS support for cross-origin requests - Real-time balance monitoring
- [Using APIMart in ChatBox](https://docs.apimart.ai/en/integrations/chat/chatbox.md): Detailed guide on how to configure and use APIMart API service in ChatBox desktop client. This guide will help you configure APIMart API in ChatBox to access rich AI model resources.
- [Using APIMart in Cherry Studio](https://docs.apimart.ai/en/integrations/chat/cherry-studio.md): Detailed guide on how to configure and use APIMart API service in Cherry Studio desktop client. This guide will help you configure APIMart API in Cherry Studio to access rich AI model resources.
- [Using APIMart in AnythingLLM](https://docs.apimart.ai/en/integrations/chat/anythingllm.md): Detailed guide on how to configure and use APIMart API service in AnythingLLM. This guide will help you configure APIMart API in AnythingLLM to build private AI knowledge bases and conversation systems.
- [Using APIMart in Claude Code](https://docs.apimart.ai/en/integrations/dev-tool/claude-code.md): A detailed guide on configuring and using APIMart API services in Claude Code CLI, enabling you to access multiple AI models for assisted programming with simple configuration.
- [Using APIMart in Cursor](https://docs.apimart.ai/en/integrations/dev-tool/cursor.md): Detailed guide on how to configure and use APIMart API service in Cursor AI code editor. This guide will help you configure APIMart API in Cursor to access rich AI model resources for code development.
- [Integrate APIMart with Cline in VSCode](https://docs.apimart.ai/en/integrations/dev-tool/cline.md): A detailed guide on how to configure APIMart API service using the Cline extension in Visual Studio Code. This guide will help you use APIMart's AI models through Cline for code development, debugging, and optimization in VSCode.
- [Using APIMart with Gemini CLI](https://docs.apimart.ai/en/integrations/dev-tool/gemini.md): Detailed guide on configuring APIMart API service in Gemini CLI command-line tool. Learn how to use APIMart’s AI models—including Gemini, GPT, and Claude series—from the terminal.
- [Using APIMart in Codex CLI](https://docs.apimart.ai/en/integrations/dev-tool/codex-cli.md): A detailed guide on configuring and using APIMart API services in Codex CLI, enabling you to access multiple AI models for assisted programming with simple configuration.
- [Using APIMart in CC-Switch](https://docs.apimart.ai/en/integrations/dev-tool/cc-switch.md): A detailed guide on configuring and using APIMart API services in CC-Switch, enabling you to access multiple AI models for assisted programming with simple configuration.
- [Using APIMart in Dify](https://docs.apimart.ai/en/integrations/platform/dify.md): Detailed guide on how to configure and use APIMart API service in Dify LLMOps platform. This guide will help you configure APIMart API in Dify to build powerful AI applications.
- [Using APIMart in Immersive Translate](https://docs.apimart.ai/en/integrations/platform/immersive-translate.md): Detailed guide on how to configure and use APIMart API service in the Immersive Translate browser extension. This guide will help you set up APIMart API in Immersive Translate for high-quality AI translation services.
- [Quick Start](https://docs.apimart.ai/en/quickstart.md): Quickly start using our API services
- [Development Guide](https://docs.apimart.ai/en/development.md): Integrate the API into your application
- [FAQs](https://docs.apimart.ai/en/faqs.md): Frequently Asked Questions about using APIMart
- [Connection & Usage](https://docs.apimart.ai/en/faqs/connection-usage.md): Troubleshoot connection and usage issues with APIMart
- [Account Management](https://docs.apimart.ai/en/faqs/account-management.md): Account usage and technical support inquiries
- [Security & Configuration](https://docs.apimart.ai/en/faqs/security-configuration.md): API key management and security configuration guide
- [Features & Capabilities](https://docs.apimart.ai/en/faqs/features.md): APIMart supported features and compatibility information
- [Cost Optimization](https://docs.apimart.ai/en/faqs/cost-optimization.md): Practical tips for optimizing API call costs

## OpenAPI Specs

- [openapi](https://docs.apimart.ai/api-reference/openapi.json)