# Rivya AI

> Rivya AI is a public AI creation workspace and developer platform. This file is a compact guide for AI agents and LLM readers; it does not replace robots.txt, sitemap.xml, canonical tags, or hreflang.

## Core public entry points

- [Rivya AI](https://rivya.ai/): Unified workspace for AI chat, image, video, audio, tools, and prompt workflows.
- [Developers](https://rivya.ai/developers): Public API overview, API keys, SDK, and integration entry.
- [API SDK docs](https://rivya.ai/docs/api-sdk): Rivya API SDK usage and integration guide.
- [Chat API docs](https://rivya.ai/docs/api-chat): Public Chat API guide, including streaming behavior.
- [Models](https://rivya.ai/ai-models): Index of public model owner pages.
- [Tools](https://rivya.ai/tools): Index of task-focused tool owner pages.
- [Docs](https://rivya.ai/docs): User and developer documentation index.

## Creation owner pages

- [AI Chat](https://rivya.ai/chat): Chat owner page and direct quick-use entry.
- [AI Image](https://rivya.ai/image): Image generation and editing owner page.
- [AI Video](https://rivya.ai/video): Video generation owner page.
- [AI Audio](https://rivya.ai/audio): Audio generation owner page.

## Model owner pages

- [Z-Image](https://rivya.ai/ai-models/z-image): Alibaba's lightweight text-to-image model. Fast single-image generation with 5 aspect ratios — ideal for quick concept drafts and social media visuals at just 1 credit.
- [Nano Banana 2](https://rivya.ai/ai-models/nano-banana-2): Nano Banana 2 is Rivya's newer Nano Banana image model for brand visuals, product images, and reference-led creative drafts when you want a more current Google-image path than the base tier.
- [Nano Banana Pro](https://rivya.ai/ai-models/nano-banana-pro): Google's premium image model with 4K output, 11 aspect ratios, and up to 8 reference images. Optimized for high-fidelity brand and campaign visuals with superior detail and color accuracy.
- [Nano Banana](https://rivya.ai/ai-models/nano-banana): Google's flexible image model for text-to-image and image-to-image with 11 aspect ratios, up to 10 reference images, and PNG/JPEG output. A strong fit for portraits, product compositions, and wider landing-page visuals.
- [GPT Image 2](https://rivya.ai/ai-models/gpt-image-2): OpenAI's newer GPT Image model on Rivya, with text-to-image, image-to-image, up to 16 reference images, and clear 1K / 2K / 4K credit tiers.
- [GPT Image 1.5](https://rivya.ai/ai-models/gpt-image-1-5): OpenAI's image model with medium/high quality tiers and up to 16 reference images. Excels at following complex instructions and rendering coherent scenes with accurate spatial relationships.
- [4o Image](https://rivya.ai/ai-models/4o-image): OpenAI's 4o Image model is now available as a dedicated text-to-image path on Rivya. It keeps the page setup intentionally narrow for now: prompt plus 3 supported aspect ratios at a fixed 3 credits per image.
- [Seedream 5.0 Lite](https://rivya.ai/ai-models/seedream-5-lite): ByteDance's lighter Seedream image model with shared pricing across text-to-image and image editing. It supports 8 aspect ratios, up to 14 reference images, and currently costs 6 credits per run.
- [Seedream 4.5](https://rivya.ai/ai-models/seedream-4-5): ByteDance's high-end image model with 2K/4K quality tiers, 8 aspect ratios, and up to 14 reference images. Known for cinematic color grading and rich texture detail in fashion and lifestyle visuals.
- [Seedream 4.0](https://rivya.ai/ai-models/seedream-4-0): Seedream 4.0 is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Seedream 3.0](https://rivya.ai/ai-models/seedream-3-0): Seedream 3.0 is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Grok Imagine](https://rivya.ai/ai-models/grok-imagine): xAI's image model with strong creative interpretation and 5 aspect ratios. Standard text requests return 2 images, while quality requests return 4.
- [Flux 2 Pro](https://rivya.ai/ai-models/flux-2-pro): Black Forest Labs' 32B-parameter flagship. Supports text-to-image and image-to-image with up to 8 reference images, 2K resolution, and accurate text rendering — built for product shots and brand visuals.
- [Flux 2 Flex](https://rivya.ai/ai-models/flux-2-flex): Flux 2 family's editing-focused variant. Specializes in structural adjustments and style transfer with up to 8 reference images and 2K resolution — ideal for iterating on existing visuals.
- [Flux Kontext Max](https://rivya.ai/ai-models/flux-kontext-max): BFL now calls FLUX.1 Kontext [max] previous-generation and recommends FLUX.2 for new projects. Black Forest Labs' enhanced Flux Kontext model for demanding prompt-led generation and image editing. Rivya offers both text-to-image and image-to-image on one model page at a fixed 8 credits per run.
- [Flux Kontext Pro](https://rivya.ai/ai-models/flux-kontext-pro): BFL now calls FLUX.1 Kontext [pro] previous-generation and recommends FLUX.2 for new projects. Black Forest Labs' lower-cost Flux Kontext model for text-to-image generation and single-image editing. Rivya offers both modes on one model page at a fixed 4 credits per run.
- [Qwen2 Image](https://rivya.ai/ai-models/qwen2-image): Alibaba's Qwen2 image model is currently integrated on Rivya as one fixed-price image project. It safely covers text-to-image and image-to-image with the shared aspect-ratio subset both public docs expose, plus PNG/JPEG output, seed reuse, and a simple NSFW switch.
- [Qwen Image](https://rivya.ai/ai-models/qwen-image): Alibaba Qwen family's image model with HD presets (Square, Portrait, Landscape) and PNG/JPEG output. Strong at Chinese-language prompts and culturally nuanced visual generation.
- [Midjourney](https://rivya.ai/ai-models/midjourney): Midjourney is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Recraft Remove Background](https://rivya.ai/ai-models/recraft-remove-background): Recraft's background-removal model on Rivya for isolating the subject from one existing image. Use it when the next step needs a transparent asset, a clean cutout, or a source image without the original background.
- [Recraft Crisp Upscale](https://rivya.ai/ai-models/recraft-crisp-upscale): Recraft's light image-upscaling model on Rivya for low-cost sharpness and clarity boosts on one approved still. Use it when the chosen image only needs a cheap polish pass before export, not a heavier delivery-grade upscale.
- [Wan 2.7 Image Pro](https://rivya.ai/ai-models/wan-2-7-image-pro): On Rivya, Wan 2.7 Image Pro currently supports text to image and image to image. It can also take up to 9 reference images. Fixed 12 credits per image; one image per request.
- [Wan 2.7 Image](https://rivya.ai/ai-models/wan-2-7-image): On Rivya, Wan 2.7 Image currently supports text to image and image to image. It can also take up to 9 reference images. Fixed 5 credits per image; one image per request.
- [Google Imagen4 Ultra](https://rivya.ai/ai-models/google-imagen4-ultra): Google Imagen4 Ultra is Rivya's premium Imagen text-to-image tier. It is currently integrated as a fixed 12-credit, single-image project with public prompt, negative prompt, aspect ratio, and seed controls.
- [Google Imagen4](https://rivya.ai/ai-models/google-imagen4): Google Imagen4 is Rivya's standard Imagen text-to-image tier. It is currently integrated as a fixed 8-credit, single-image project with public prompt, negative prompt, aspect ratio, and seed controls.
- [Google Imagen4 Fast](https://rivya.ai/ai-models/google-imagen4-fast): Google Imagen4 Fast is Rivya's lightweight Imagen text-to-image tier. It currently keeps a single-image project, uses fixed 4-credit pricing, and exposes the public prompt, negative prompt, aspect ratio, and seed controls without opening multi-image output.
- [Topaz Image Upscaler](https://rivya.ai/ai-models/topaz-image-upscale): Topaz's delivery-grade image upscaler on Rivya for approved stills that need a real size jump. Use it when the composition is already final and the remaining problem is export resolution, review size, or print readiness.
- [Ideogram V3](https://rivya.ai/ai-models/ideogram-v3): Ideogram V3 is Rivya's text-to-image model for text rendering, poster layouts, and design-first image prompts. Current pricing is 4 credits for TURBO, 7 for BALANCED, and 10 for QUALITY.
- [Ideogram V3 Reframe](https://rivya.ai/ai-models/ideogram-v3-reframe): Ideogram V3 Reframe is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Ideogram V3 Remix](https://rivya.ai/ai-models/ideogram-v3-remix): Ideogram V3 Remix: TURBO 4 / BALANCED 7 / QUALITY 10 credits per image. Choose 1–4 outputs. The total is the per-image rate multiplied by the selected count, rounded up to whole credits.
- [Ideogram Character](https://rivya.ai/ai-models/ideogram-character): Character-consistency option for turning one approved character image into new scenes, outfits, and formats. TURBO 12 / BALANCED 18 / QUALITY 24 credits per image. Choose 1–4 outputs. The total is the per-image rate multiplied by the selected count, rounded up to whole credits.
- [Seedream 5.0 Pro](https://rivya.ai/ai-models/seedream-5-pro): Seedream 5.0 Pro is Rivya's multi-reference image model for campaign frames, editorial scenes, and product compositions, with text-to-image, image-to-image, 1K or 2K output, and up to 10 reference images.
- [Seedream 5.0 Pro Layer Decomposition](https://rivya.ai/ai-models/seedream-5-pro-layer-decomposition): Seedream 5.0 Pro Layer Decomposition separates one source image into a base image and reusable PNG layers, with optional prompt or bounding-box guidance and Auto, 1K, 1.5K, or 2K sizing.
- [Nano Banana 2 Lite](https://rivya.ai/ai-models/nano-banana-2-lite): Nano Banana 2 Lite is Rivya's lower-credit Nano Banana option for quick product concepts, social image variants, and reference-led drafts, with a fixed 3-credit generation price and 15 aspect-ratio choices.
- [Qwen Image 3.0](https://rivya.ai/ai-models/qwen-image-3): Qwen Image 3.0 is Rivya's configurable Qwen image model for Chinese or English prompts, reference-led layouts, and reproducible visual tests with 1K or 2K output, prompt expansion, negative prompts, and seed control.
- [Qwen Image 3.0 Pro](https://rivya.ai/ai-models/qwen-image-3-pro): Qwen Image 3.0 Pro is Rivya's separate Pro-tier Qwen image model for polished campaign key visuals, product scenes, and poster concepts, with Chinese or English prompts, up to three references, and distinct 1K and 2K credit tiers.
- [Grok Imagine Image 2.0](https://rivya.ai/ai-models/grok-imagine-image-2-0): Grok Imagine Image 2.0 supports single-image text generation and standard image editing with one to five signed reference images, five fixed aspect ratios, and an automatic ratio option for edit requests.
- [Seedance 2.0](https://rivya.ai/ai-models/seedance-2): ByteDance's full Seedance 2.0 video model with explicit support for prompt-only generation, frame-driven animation, and multimodal reference generation. Rivya keeps the documented role split explicit so frame inputs and multimodal references stay mutually exclusive instead of collapsing into one ambiguous upload bucket.
- [Seedance 2.0 Fast](https://rivya.ai/ai-models/seedance-2-fast): ByteDance's faster Seedance 2.0 video model with full scene routing for prompt-only generation, frame-driven image animation, and multimodal reference video generation. Rivya keeps the documented scene split explicit so first/last-frame inputs do not collide with reference image, video, and audio roles.
- [Seedance 1.5 Pro](https://rivya.ai/ai-models/seedance-1-5-pro): A Seedance video model for text-to-video and image-to-video with native audio-visual sync. It supports 480p–1080p, 4–12s clips, 6 aspect ratios, dynamic or fixed lens control, optional audio generation, and lip-sync.
- [Seedance 1.0 Pro](https://rivya.ai/ai-models/seedance-1-0-pro): Seedance 1.0 Pro is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Seedance 1.0 Pro Fast](https://rivya.ai/ai-models/seedance-1-0-pro-fast): Seedance 1.0 Pro Fast is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Seedance 1.0 Lite](https://rivya.ai/ai-models/seedance-1-0-lite): Seedance 1.0 Lite is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [HappyHorse 1.0](https://rivya.ai/ai-models/happyhorse-1-0): HappyHorse 1.0 on Rivya supports text-to-video, single-image animation, and multi-image reference video from one model page.
- [Wan 2.7 Video](https://rivya.ai/ai-models/wan-2-7-video): Wan 2.7 Video: 720p = 16 credits/s; 1080p = 24 credits/s; text/image: integer 2–15 s; video edit: duration=0 or integer 2–10 s; nsfw_checker=true.
- [Wan 2.6](https://rivya.ai/ai-models/wan-2-6): Wan 2.6 supports text-to-video with no media, image-to-video with exactly one image, and video editing with one to three videos; images and videos cannot be mixed. Text and image inputs support 5, 10, or 15 seconds; video input supports 5 or 10 seconds only.
- [Wan 2.5 Video](https://rivya.ai/ai-models/wan-2-5-video): Wan 2.5 is now exposed on Rivya as one shared entry for text-to-video and image-to-video. Current pricing is `720p_5 = 60`, `720p_10 = 120`, `1080p_5 = 100`, and `1080p_10 = 200` credits.
- [Wan 2.2 A14B Turbo](https://rivya.ai/ai-models/wan-2-2-a14b-turbo): Wan 2.2 A14B Turbo covers video from text, image, and image plus audio. Text or image runs are fixed at 5 seconds and cost `480p = 40`, `580p = 60`, `720p = 80`; image-plus-audio runs cost `480p = 12`, `580p = 18`, `720p = 24` per output second.
- [Wan Animate Replace](https://rivya.ai/ai-models/wan-2-2-animate-replace): Choose Wan Animate Replace when the source clip already has the right motion and framing, and only the visible subject should change. Go back to prompt-led generation when the whole shot needs to be reimagined, or use a broader editing project when more than a subject swap is on the table.
- [Hailuo 2.3](https://rivya.ai/ai-models/hailuo-2-3): MiniMax's image-to-video model with Standard/Pro quality tiers, 768P/1080P resolution, and 6s or 10s clips. Known for smoother motion and natural transitions from still images.
- [Hailuo Pro](https://rivya.ai/ai-models/hailuo-pro): MiniMax's current Hailuo Pro video model is connected here as one fixed Pro-tier model for both text-to-video and image-to-video. Image mode accepts 1 or 2 reference images, with the second image used as the last frame, and each run currently costs 57 credits. Fixed output: 1080P · 6s · 57 credits.
- [Hailuo Standard](https://rivya.ai/ai-models/hailuo-standard): MiniMax's older Hailuo Standard video model, unified here as one model for both text-to-video and image-to-video. Image mode accepts 1 or 2 reference images, with the second image used as the last frame, and the currently verified public pricing tiers range from 12 to 50 credits.
- [Kling 3.0](https://rivya.ai/ai-models/kling-3-0): Kuaishou's premium video model for text-to-video and image-to-video, with Standard (720P), Pro (1080P), and 4K tiers, single or multi-shot structure, 3–15s duration, optional audio generation, and up to 2 reference images.
- [Kling 3.0 motion-control](https://rivya.ai/ai-models/kling-3-0-motion-control): Exact 1-image + 1-motion-video project keeps identity and movement roles clear Standard (720P): 20 × verified motion-video seconds; Pro (1080P): 27 × seconds
- [Kling 2.6](https://rivya.ai/ai-models/kling-2-6): Kuaishou's video model with optional audio generation, 5s/10s clips, and 3 aspect ratios. Strong at human motion and expressive character animation with natural physics.
- [Kling 2.6 motion-control](https://rivya.ai/ai-models/kling-2-6-motion-control): Dedicated motion-transfer project for driving one subject from one reference image plus one motion video. Use it when you want a cheaper Kling motion-control pass and can live without the extra scene controls in Kling 3.0 motion-control.
- [Kling V2.5 Turbo Pro](https://rivya.ai/ai-models/kling-v2-5-turbo-pro): Kuaishou's Kling V2.5 Turbo Pro video model, now supporting both text-to-video and image-to-video. Public pricing evidence clearly covers both text and image tiers at 5 seconds and 10 seconds, so Rivya maps it directly to 42 / 84 credits.
- [Kling V2.1 Master](https://rivya.ai/ai-models/kling-v2-1-master): Kuaishou's older Kling V2.1 Master video model now supports both text-to-video and image-to-video on Rivya. Current pricing is 160 credits for 5 seconds and 320 credits for 10 seconds.
- [Kling V2.1 Pro](https://rivya.ai/ai-models/kling-v2-1-pro): Kuaishou's older Kling V2.1 Pro image-to-video model supports a first frame plus an optional tail frame image. Current pricing is 50 credits for 5 seconds and 100 credits for 10 seconds.
- [Kling V2.1 Standard](https://rivya.ai/ai-models/kling-v2-1-standard): Kuaishou's older Kling V2.1 Standard image-to-video model. Current pricing is 25 credits for 5 seconds and 50 credits for 10 seconds.
- [Kling AI Avatar Pro](https://rivya.ai/ai-models/kling-ai-avatar-pro): Kuaishou's Kling AI Avatar Pro higher-quality talking-avatar model, using one portrait image plus one audio clip to generate lip-synced avatar video. Rivya charges 16 credits per second of verified audio, up to 300 seconds, with the total rounded up to a whole credit.
- [Kling AI Avatar Standard](https://rivya.ai/ai-models/kling-ai-avatar-standard): Kuaishou's Kling AI Avatar Standard talking-avatar model, using one portrait image plus one audio clip to generate lip-synced avatar video. Rivya charges 8 credits per second of verified audio, up to 300 seconds, with the total rounded up to a whole credit.
- [Infinitalk](https://rivya.ai/ai-models/infinitalk): Infinitalk is a portrait-plus-audio talking-video model. Current pricing is metered by resolution and audio duration: 480p = 3 credits per second and 720p = 12 credits per second.
- [Runway](https://rivya.ai/ai-models/runway): Runway is a standalone video model that supports both text-to-video and image-to-video. Public pricing evidence currently confirms only 3 generation tiers, so Rivya keeps it on the verified set: `720p_5 = 12`, `720p_10 = 30`, and `1080p_5 = 30`.
- [Runway Aleph](https://rivya.ai/ai-models/runway-aleph): Source-video transformation project for reworking an existing clip into a new visual result. Use Aleph when the motion comes from your input footage and the creative direction comes from your prompt, with a fixed 90-credit price.
- [Luma Modify Video](https://rivya.ai/ai-models/luma-modify-video): Luma Modify Video is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Grok Imagine Video 1.5 Preview](https://rivya.ai/ai-models/grok-imagine-video): Grok Imagine Video 1.5 Preview supports 1–15 seconds, 480p or 720p, up to 7 reference images, and prompts up to 4096 characters.
- [Sora 2 Pro](https://rivya.ai/ai-models/sora-2-pro): Sora 2 Pro is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Sora 2](https://rivya.ai/ai-models/sora-2): Sora 2 is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Sora Watermark Remover](https://rivya.ai/ai-models/sora-watermark-remover): Sora Watermark Remover is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Topaz Video Upscaler](https://rivya.ai/ai-models/topaz-video-upscale): One uploaded video, no prompt, with a 1x, 2x, or 4x upscale factor; 2x is the default.
- [Veo3.1 Quality](https://rivya.ai/ai-models/veo3-1-quality): 720p costs 225 credits, 1080p costs 233 (232.5 rounded up), and 4K costs 285; Extend costs 225, while eligible 1080p/4K result actions cost 8/60. Audio is experimental and not guaranteed; review every result manually before delivery. Resolution actions require an owned successful 720p base task or Extend result; a direct 1080p/4K result cannot request another resolution action.
- [Veo3.1 Fast](https://rivya.ai/ai-models/veo3-1-fast): Supports text, first/last-frame, and one-to-three-reference-image generation with direct 720p/1080p/4K and 4/6/8-second controls. Audio is experimental and not guaranteed; review every result manually before delivery.
- [Veo3.1 Lite](https://rivya.ai/ai-models/veo3-1-lite): Supports text, first/last-frame, and one-to-three-reference-image generation with direct 720p/1080p/4K and 4/6/8-second controls. Audio is experimental and not guaranteed; review every result manually before delivery.
- [Wan 3.0 Video](https://rivya.ai/ai-models/wan-3-0-video): Wan 3.0 Video combines Standard and Prime tiers with text-to-video, first/last-frame guidance, and signed multimodal references. It supports 480P, 720P, or 1080P output, explicit 2-30-second clips or intelligent duration, and reference bundles containing images, video, and audio.
- [Flux 3](https://rivya.ai/ai-models/flux-3): Flux 3 appears in the coming-soon video catalog with Text to Video and Image to Video labels. Rivya keeps this page as a status reference, not as a promise of controls or runtime behavior.
- [Kling 3.0 Turbo](https://rivya.ai/ai-models/kling-3-turbo): Kling 3.0 Turbo is Rivya's focused Kling video model for prompt-only generation or animation from one still image. Its two paths share 720p and 1080p output with integer durations from 3 to 15 seconds, while aspect-ratio control stays exclusive to text-to-video.
- [Seedance 2.0 Mini](https://rivya.ai/ai-models/seedance-2-mini): Seedance 2.0 Mini is Rivya's lower-rate multimodal Seedance workflow for prompt-only video, first- or first-and-last-frame animation, and mixed image, video, or audio references. Its 480p and 720p tiers cover 4-to-15-second output while keeping the three input scenes mutually exclusive.
- [Seedance 2.5](https://rivya.ai/ai-models/seedance-2-5): Seedance 2.5 is Rivya's larger Seedance workflow for prompt-only video, first- and last-frame guidance, multimodal references, and video-guided transformation. It adds a 1080p tier, output up to 30 seconds, automatic duration, MP4 or MOV output, and substantially larger image, video, and audio reference capacity than Mini.
- [MiniMax H3](https://rivya.ai/ai-models/minimax-h3): MiniMax H3 is Rivya's 768P or 2K video model for prompt-only generation, first- and last-frame animation, and multimodal reference work. Its reference path can combine images, MP4 or MOV clips, and MP3 or WAV audio, while its credit formula separately accounts for verified input-video time and images beyond the first five.
- [HappyHorse 1.1](https://rivya.ai/ai-models/happyhorse-1-1): HappyHorse 1.1 is a flexible AI video model for prompt-only clips, single-image animation, and multi-image reference video at 720p or 1080p, with 3 to 15 seconds of output.
- [Gemini Omni Video](https://rivya.ai/ai-models/gemini-omni-video): The official contract remains documented, but the current product page is unavailable and the signed-in pricing catalog has no price for this model. Generation is therefore blocked before credit reservation.
- [OmniHuman 1.5](https://rivya.ai/ai-models/omnihuman-1-5): OmniHuman 1.5 turns exactly one portrait image and one audio clip into an audio-driven character video at 720p or 1080p, with credits based on verified audio duration.
- [Volcengine Video Lip Sync](https://rivya.ai/ai-models/volcengine-video-lip-sync): Volcengine Video Lip Sync replaces the spoken performance in one existing video with one target vocal track, using Lite or Basic processing and duration-based credits. The output is MP4 at 25 FPS, and its duration follows the target audio.
- [Kling O3](https://rivya.ai/ai-models/kling-o3): Verified single-shot generation is available from text, one first frame, or a first-and-last-frame pair.
- [Wan Animate 2](https://rivya.ai/ai-models/wan-animate-2): Wan Animate 2 is a Coming Soon Video to Video listing. Its public character-animation positioning does not provide a callable contract or price for Rivya.
- [PixVerse V6](https://rivya.ai/ai-models/pixverse-v6): PixVerse V6 — Temporarily unavailable. This is a temporary pricing and integration-safety hold, not a Coming Soon announcement. The page remains indexable, but generation and callable pricing stay closed until the complete price contract is stable.
- [Wan 2.6 Flash](https://rivya.ai/ai-models/wan-2-6-flash): Wan 2.6 Flash / Temporarily unavailable — This is a temporary pricing and integration-safety hold, not a Coming Soon announcement. The page remains indexable, but generation and callable pricing stay closed until the complete price contract is stable.
- [Suno Music](https://rivya.ai/ai-models/suno-music): Suno Music is Rivya's text-to-music model for turning one short brief into a first song draft with or without vocals. It keeps the fixed `12` credit entry point and exposes `Extend Music` as the next step after a successful track.
- [Suno Sounds](https://rivya.ai/ai-models/suno-sounds): Suno Sounds is Rivya's lightweight text-to-sound model for ambience loops, background sound, and short sonic sketches. It keeps the documented fixed price of `2.5` credits per generation and lets successful results continue into `Vocal Separation`.
- [Suno Lyrics](https://rivya.ai/ai-models/suno-lyrics): Suno Lyrics is Rivya's lyric-generation model for turning one theme or mood into song words at a fixed cost of `1` credit per request.
- [ElevenLabs Dialogue V3](https://rivya.ai/ai-models/elevenlabs-dialogue-v3): ElevenLabs' multi-speaker dialogue model on Rivya. It is built for role-based speech generation, with individual voice assignments, stability controls, and dialogue-ready pacing for podcasts, interviews, and character scenes.
- [ElevenLabs Turbo 2.5](https://rivya.ai/ai-models/elevenlabs-turbo-2-5): ElevenLabs' fast text-to-speech model on Rivya. With low-latency voice generation and adjustable stability, similarity, style, and speed, it is built for rapid voiceover drafts and interactive TTS projects.
- [ElevenLabs Multilingual V2](https://rivya.ai/ai-models/elevenlabs-multilingual-v2): ElevenLabs' multilingual text-to-speech model on Rivya, supporting about 30 languages with auto-detection. It is the stronger option for localization, cross-language delivery, and more natural multilingual voiceovers.
- [ElevenLabs Sound Effect V2](https://rivya.ai/ai-models/elevenlabs-sound-effect-v2): ElevenLabs Sound Effect V2 is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [ElevenLabs Audio Isolation](https://rivya.ai/ai-models/elevenlabs-audio-isolation): ElevenLabs Audio Isolation is currently unavailable on Rivya. This page is a historical capability reference only; it offers no generation, creates no task, and deducts no credits.
- [Gemini 3.1 Flash TTS](https://rivya.ai/ai-models/gemini-3-1-flash-tts): Gemini 3.1 Flash TTS is listed for text-to-speech and dialogue-to-speech work, but it is temporarily unavailable on Rivya because a run's maximum audio-token cost cannot yet be bounded safely before it starts.
- [Gemini 2.5 Pro TTS](https://rivya.ai/ai-models/gemini-2-5-pro-tts): Gemini 2.5 Pro TTS has a public text-to-speech and dialogue-to-speech listing, while Rivya keeps generation temporarily unavailable until every allowed script can be covered by a safe maximum credit reservation.
- [GPT-5.5](https://rivya.ai/ai-models/gpt-5-5-chat): GPT-5.5 is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5.4](https://rivya.ai/ai-models/gpt-5-4-chat): GPT-5.4 is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5.4 Codex](https://rivya.ai/ai-models/gpt-5-4-codex-chat): GPT-5.4 Codex is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5.3 Codex](https://rivya.ai/ai-models/gpt-5-3-codex-chat): GPT-5.3 Codex is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5.2](https://rivya.ai/ai-models/gpt-5-2-chat): GPT-5.2 is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5.2 Codex](https://rivya.ai/ai-models/gpt-5-2-codex-chat): GPT-5.2 Codex is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5.1 Codex](https://rivya.ai/ai-models/gpt-5-1-codex-chat): GPT-5.1 Codex is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5 Codex](https://rivya.ai/ai-models/gpt-5-codex-chat): GPT-5 Codex is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Gemini 3.1 Pro](https://rivya.ai/ai-models/gemini-3-1-pro-chat): Gemini 3.1 Pro is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Gemini 3 Pro](https://rivya.ai/ai-models/gemini-3-pro-chat): Gemini 3 Pro is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Gemini 3 Flash](https://rivya.ai/ai-models/gemini-3-flash-chat): Gemini 3 Flash is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Gemini 2.5 Pro](https://rivya.ai/ai-models/gemini-2-5-pro-chat): Gemini 2.5 Pro is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Gemini 2.5 Flash](https://rivya.ai/ai-models/gemini-2-5-flash-chat): Gemini 2.5 Flash is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Claude Opus 4.7](https://rivya.ai/ai-models/claude-opus-4-7-chat): Anthropic's flagship Claude chat model on Rivya for deep reasoning, careful synthesis, executive writing, and high-impact text work.
- [Claude Opus 4.6](https://rivya.ai/ai-models/claude-opus-4-6-chat): Claude Opus 4.6 is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Claude Sonnet 4.6](https://rivya.ai/ai-models/claude-sonnet-4-6-chat): Anthropic's balanced Claude AI chat model on Rivya. It keeps strong long-form reasoning and careful analysis for content, research, and coding projects without jumping to Opus-level spend.
- [Claude Opus 4.5](https://rivya.ai/ai-models/claude-opus-4-5-chat): Claude Opus 4.5 is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Claude Sonnet 4.5](https://rivya.ai/ai-models/claude-sonnet-4-5-chat): Anthropic's balanced Claude AI chat model on Rivya. It is strong at nuanced writing, careful analysis, and safety-conscious responses, making it a strong Claude option for content creation and research.
- [Claude Haiku 4.5](https://rivya.ai/ai-models/claude-haiku-4-5-chat): Anthropic's lightweight Claude AI chat model on Rivya. It is tuned for speed, cost efficiency, and stable day-to-day chat performance in high-frequency projects where you want Claude tone without premium spend.
- [Claude Opus 4.8](https://rivya.ai/ai-models/claude-opus-4-8-chat): A text-first Claude chat model on Rivya for reviewing dense source packets and producing careful, structured decisions.
- [Claude Fable 5](https://rivya.ai/ai-models/claude-fable-5-chat): A text-first Claude chat model on Rivya for narrative reframing, audience-aware drafts, and structured editorial work.
- [Claude Sonnet 5](https://rivya.ai/ai-models/claude-sonnet-5-chat): A text-first Claude chat model on Rivya for implementation briefs, requirements review, and concise technical planning.
- [Claude Opus 5](https://rivya.ai/ai-models/claude-opus-5-chat): A text-first Claude chat model on Rivya for board-level risk framing, scenario comparison, and decision documentation.
- [GPT-5.6 Luna](https://rivya.ai/ai-models/gpt-5-6-luna-chat): GPT-5.6 Luna is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5.6 Terra](https://rivya.ai/ai-models/gpt-5-6-terra-chat): GPT-5.6 Terra is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [GPT-5.6 Sol](https://rivya.ai/ai-models/gpt-5-6-sol-chat): GPT-5.6 Sol is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Gemini 3.5 Flash](https://rivya.ai/ai-models/gemini-3-5-flash-chat): Gemini 3.5 Flash is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Gemini 3.6 Flash](https://rivya.ai/ai-models/gemini-3-6-flash-chat): Gemini 3.6 Flash is listed on Rivya with an integration-safety hold. Its indexable status page remains available while chat execution stays disabled.
- [Gemini 3.7 Flash](https://rivya.ai/ai-models/gemini-3-7-flash-chat): Gemini 3.7 Flash is listed on Rivya with an integration-safety hold. Its indexable status page tracks the confirmed chat contract while execution stays disabled.
- [Grok 4.3](https://rivya.ai/ai-models/grok-4-3-chat): Grok 4.3 — Temporarily unavailable. Continue with a callable alternative while Grok 4.3 remains on integration-safety hold.
- [Grok 4.5](https://rivya.ai/ai-models/grok-4-5-chat): Grok 4.5 — Temporarily unavailable. Continue with a callable alternative while Grok 4.5 remains on integration-safety hold.
- [Grok 4.6](https://rivya.ai/ai-models/grok-4-6-chat): Grok 4.6 — Temporarily unavailable. Continue with a callable alternative while Grok 4.6 remains on integration-safety hold.
- [Kimi K3](https://rivya.ai/ai-models/kimi-k3-chat): Kimi K3 is a coming-soon Chat model. Rivya uses this page to track availability and direct visitors to available chat models without claiming unpublished context, tools, or response behavior.

## Tool owner pages

- [AI Calculator](https://rivya.ai/tools/ai-calculator): Use natural-language prompts for business math, formulas, unit conversions, percentages, growth, margin, pricing, and scenario comparison with the final answer first and checkable steps.
- [AI Solver](https://rivya.ai/tools/ai-solver): Solve equations, word problems, homework-style questions, reasoning tasks, and concept explanations with the final answer first and clear guided steps after.

## Public API model docs

- [Z-Image API](https://rivya.ai/docs/api/models/z-image): Alibaba's lightweight text-to-image model. Fast single-image generation with 5 aspect ratios — ideal for quick concept drafts and social media visuals at just 1 credit.
- [Nano Banana 2 API](https://rivya.ai/docs/api/models/nano-banana-2): Nano Banana 2 is Rivya's newer Nano Banana image model for brand visuals, product images, and reference-led creative drafts when you want a more current Google-image path than the base tier.
- [Nano Banana Pro API](https://rivya.ai/docs/api/models/nano-banana-pro): Google's premium image model with 4K output, 11 aspect ratios, and up to 8 reference images. Optimized for high-fidelity brand and campaign visuals with superior detail and color accuracy.
- [Nano Banana API](https://rivya.ai/docs/api/models/nano-banana): Google's flexible image model for text-to-image and image-to-image with 11 aspect ratios, up to 10 reference images, and PNG/JPEG output. A strong fit for portraits, product compositions, and wider landing-page visuals.
- [GPT Image 2 API](https://rivya.ai/docs/api/models/gpt-image-2): OpenAI's newer GPT Image model on Rivya, with text-to-image, image-to-image, up to 16 reference images, and clear 1K / 2K / 4K credit tiers.
- [GPT Image 1.5 API](https://rivya.ai/docs/api/models/gpt-image-1-5): OpenAI's image model with medium/high quality tiers and up to 16 reference images. Excels at following complex instructions and rendering coherent scenes with accurate spatial relationships.
- [4o Image API](https://rivya.ai/docs/api/models/4o-image): OpenAI's 4o Image model is now available as a dedicated text-to-image path on Rivya. It keeps the page setup intentionally narrow for now: prompt plus 3 supported aspect ratios at a fixed 3 credits per image.
- [Seedream 5.0 Lite API](https://rivya.ai/docs/api/models/seedream-5-lite): ByteDance's lighter Seedream image model with shared pricing across text-to-image and image editing. It supports 8 aspect ratios, up to 14 reference images, and currently costs 6 credits per run.
- [Seedream 4.5 API](https://rivya.ai/docs/api/models/seedream-4-5): ByteDance's high-end image model with 2K/4K quality tiers, 8 aspect ratios, and up to 14 reference images. Known for cinematic color grading and rich texture detail in fashion and lifestyle visuals.
- [Grok Imagine API](https://rivya.ai/docs/api/models/grok-imagine): xAI's image model with strong creative interpretation and 5 aspect ratios. Standard text requests return 2 images, while quality requests return 4.
- [Flux 2 Pro API](https://rivya.ai/docs/api/models/flux-2-pro): Black Forest Labs' 32B-parameter flagship. Supports text-to-image and image-to-image with up to 8 reference images, 2K resolution, and accurate text rendering — built for product shots and brand visuals.
- [Flux 2 Flex API](https://rivya.ai/docs/api/models/flux-2-flex): Flux 2 family's editing-focused variant. Specializes in structural adjustments and style transfer with up to 8 reference images and 2K resolution — ideal for iterating on existing visuals.
- [Flux Kontext Max API](https://rivya.ai/docs/api/models/flux-kontext-max): BFL now calls FLUX.1 Kontext [max] previous-generation and recommends FLUX.2 for new projects. Black Forest Labs' enhanced Flux Kontext model for demanding prompt-led generation and image editing. Rivya offers both text-to-image and image-to-image on one model page at a fixed 8 credits per run.
- [Flux Kontext Pro API](https://rivya.ai/docs/api/models/flux-kontext-pro): BFL now calls FLUX.1 Kontext [pro] previous-generation and recommends FLUX.2 for new projects. Black Forest Labs' lower-cost Flux Kontext model for text-to-image generation and single-image editing. Rivya offers both modes on one model page at a fixed 4 credits per run.
- [Qwen2 Image API](https://rivya.ai/docs/api/models/qwen2-image): Alibaba's Qwen2 image model is currently integrated on Rivya as one fixed-price image project. It safely covers text-to-image and image-to-image with the shared aspect-ratio subset both public docs expose, plus PNG/JPEG output, seed reuse, and a simple NSFW switch.
- [Qwen Image API](https://rivya.ai/docs/api/models/qwen-image): Alibaba Qwen family's image model with HD presets (Square, Portrait, Landscape) and PNG/JPEG output. Strong at Chinese-language prompts and culturally nuanced visual generation.
- [Recraft Remove Background API](https://rivya.ai/docs/api/models/recraft-remove-background): Recraft's background-removal model on Rivya for isolating the subject from one existing image. Use it when the next step needs a transparent asset, a clean cutout, or a source image without the original background.
- [Recraft Crisp Upscale API](https://rivya.ai/docs/api/models/recraft-crisp-upscale): Recraft's light image-upscaling model on Rivya for low-cost sharpness and clarity boosts on one approved still. Use it when the chosen image only needs a cheap polish pass before export, not a heavier delivery-grade upscale.
- [Wan 2.7 Image Pro API](https://rivya.ai/docs/api/models/wan-2-7-image-pro): On Rivya, Wan 2.7 Image Pro currently supports text to image and image to image. It can also take up to 9 reference images. Fixed 12 credits per image; one image per request.
- [Wan 2.7 Image API](https://rivya.ai/docs/api/models/wan-2-7-image): On Rivya, Wan 2.7 Image currently supports text to image and image to image. It can also take up to 9 reference images. Fixed 5 credits per image; one image per request.
- [Google Imagen4 Ultra API](https://rivya.ai/docs/api/models/google-imagen4-ultra): Google Imagen4 Ultra is Rivya's premium Imagen text-to-image tier. It is currently integrated as a fixed 12-credit, single-image project with public prompt, negative prompt, aspect ratio, and seed controls.
- [Google Imagen4 API](https://rivya.ai/docs/api/models/google-imagen4): Google Imagen4 is Rivya's standard Imagen text-to-image tier. It is currently integrated as a fixed 8-credit, single-image project with public prompt, negative prompt, aspect ratio, and seed controls.
- [Google Imagen4 Fast API](https://rivya.ai/docs/api/models/google-imagen4-fast): Google Imagen4 Fast is Rivya's lightweight Imagen text-to-image tier. It currently keeps a single-image project, uses fixed 4-credit pricing, and exposes the public prompt, negative prompt, aspect ratio, and seed controls without opening multi-image output.
- [Topaz Image Upscaler API](https://rivya.ai/docs/api/models/topaz-image-upscale): Topaz's delivery-grade image upscaler on Rivya for approved stills that need a real size jump. Use it when the composition is already final and the remaining problem is export resolution, review size, or print readiness.
- [Ideogram V3 API](https://rivya.ai/docs/api/models/ideogram-v3): Ideogram V3 is Rivya's text-to-image model for text rendering, poster layouts, and design-first image prompts. Current pricing is 4 credits for TURBO, 7 for BALANCED, and 10 for QUALITY.
- [Ideogram V3 Remix API](https://rivya.ai/docs/api/models/ideogram-v3-remix): Ideogram V3 Remix: TURBO 4 / BALANCED 7 / QUALITY 10 credits per image. Choose 1–4 outputs. The total is the per-image rate multiplied by the selected count, rounded up to whole credits.
- [Ideogram Character API](https://rivya.ai/docs/api/models/ideogram-character): Character-consistency option for turning one approved character image into new scenes, outfits, and formats. TURBO 12 / BALANCED 18 / QUALITY 24 credits per image. Choose 1–4 outputs. The total is the per-image rate multiplied by the selected count, rounded up to whole credits.
- [Seedream 5.0 Pro API](https://rivya.ai/docs/api/models/seedream-5-pro): Seedream 5.0 Pro is Rivya's multi-reference image model for campaign frames, editorial scenes, and product compositions, with text-to-image, image-to-image, 1K or 2K output, and up to 10 reference images.
- [Seedream 5.0 Pro Layer Decomposition API](https://rivya.ai/docs/api/models/seedream-5-pro-layer-decomposition): Seedream 5.0 Pro Layer Decomposition separates one source image into a base image and reusable PNG layers, with optional prompt or bounding-box guidance and Auto, 1K, 1.5K, or 2K sizing.
- [Nano Banana 2 Lite API](https://rivya.ai/docs/api/models/nano-banana-2-lite): Nano Banana 2 Lite is Rivya's lower-credit Nano Banana option for quick product concepts, social image variants, and reference-led drafts, with a fixed 3-credit generation price and 15 aspect-ratio choices.
- [Qwen Image 3.0 API](https://rivya.ai/docs/api/models/qwen-image-3): Qwen Image 3.0 is Rivya's configurable Qwen image model for Chinese or English prompts, reference-led layouts, and reproducible visual tests with 1K or 2K output, prompt expansion, negative prompts, and seed control.
- [Qwen Image 3.0 Pro API](https://rivya.ai/docs/api/models/qwen-image-3-pro): Qwen Image 3.0 Pro is Rivya's separate Pro-tier Qwen image model for polished campaign key visuals, product scenes, and poster concepts, with Chinese or English prompts, up to three references, and distinct 1K and 2K credit tiers.
- [Grok Imagine Image 2.0 API](https://rivya.ai/docs/api/models/grok-imagine-image-2-0): Grok Imagine Image 2.0 supports single-image text generation and standard image editing with one to five signed reference images, five fixed aspect ratios, and an automatic ratio option for edit requests.
- [Seedance 2.0 API](https://rivya.ai/docs/api/models/seedance-2): ByteDance's full Seedance 2.0 video model with explicit support for prompt-only generation, frame-driven animation, and multimodal reference generation. Rivya keeps the documented role split explicit so frame inputs and multimodal references stay mutually exclusive instead of collapsing into one ambiguous upload bucket.
- [Seedance 2.0 Fast API](https://rivya.ai/docs/api/models/seedance-2-fast): ByteDance's faster Seedance 2.0 video model with full scene routing for prompt-only generation, frame-driven image animation, and multimodal reference video generation. Rivya keeps the documented scene split explicit so first/last-frame inputs do not collide with reference image, video, and audio roles.
- [Seedance 1.5 Pro API](https://rivya.ai/docs/api/models/seedance-1-5-pro): A Seedance video model for text-to-video and image-to-video with native audio-visual sync. It supports 480p–1080p, 4–12s clips, 6 aspect ratios, dynamic or fixed lens control, optional audio generation, and lip-sync.
- [HappyHorse 1.0 API](https://rivya.ai/docs/api/models/happyhorse-1-0): HappyHorse 1.0 on Rivya supports text-to-video, single-image animation, and multi-image reference video from one model page.
- [Wan 2.7 Video API](https://rivya.ai/docs/api/models/wan-2-7-video): Wan 2.7 Video: 720p = 16 credits/s; 1080p = 24 credits/s; text/image: integer 2–15 s; video edit: duration=0 or integer 2–10 s; nsfw_checker=true.
- [Wan 2.6 API](https://rivya.ai/docs/api/models/wan-2-6): Wan 2.6 supports text-to-video with no media, image-to-video with exactly one image, and video editing with one to three videos; images and videos cannot be mixed. Text and image inputs support 5, 10, or 15 seconds; video input supports 5 or 10 seconds only.
- [Wan 2.5 Video API](https://rivya.ai/docs/api/models/wan-2-5-video): Wan 2.5 is now exposed on Rivya as one shared entry for text-to-video and image-to-video. Current pricing is `720p_5 = 60`, `720p_10 = 120`, `1080p_5 = 100`, and `1080p_10 = 200` credits.
- [Wan 2.2 A14B Turbo API](https://rivya.ai/docs/api/models/wan-2-2-a14b-turbo): Wan 2.2 A14B Turbo covers video from text, image, and image plus audio. Text or image runs are fixed at 5 seconds and cost `480p = 40`, `580p = 60`, `720p = 80`; image-plus-audio runs cost `480p = 12`, `580p = 18`, `720p = 24` per output second.
- [Wan Animate Replace API](https://rivya.ai/docs/api/models/wan-2-2-animate-replace): Choose Wan Animate Replace when the source clip already has the right motion and framing, and only the visible subject should change. Go back to prompt-led generation when the whole shot needs to be reimagined, or use a broader editing project when more than a subject swap is on the table.
- [Hailuo 2.3 API](https://rivya.ai/docs/api/models/hailuo-2-3): MiniMax's image-to-video model with Standard/Pro quality tiers, 768P/1080P resolution, and 6s or 10s clips. Known for smoother motion and natural transitions from still images.
- [Hailuo Pro API](https://rivya.ai/docs/api/models/hailuo-pro): MiniMax's current Hailuo Pro video model is connected here as one fixed Pro-tier model for both text-to-video and image-to-video. Image mode accepts 1 or 2 reference images, with the second image used as the last frame, and each run currently costs 57 credits. Fixed output: 1080P · 6s · 57 credits.
- [Hailuo Standard API](https://rivya.ai/docs/api/models/hailuo-standard): MiniMax's older Hailuo Standard video model, unified here as one model for both text-to-video and image-to-video. Image mode accepts 1 or 2 reference images, with the second image used as the last frame, and the currently verified public pricing tiers range from 12 to 50 credits.
- [Kling 3.0 API](https://rivya.ai/docs/api/models/kling-3-0): Kuaishou's premium video model for text-to-video and image-to-video, with Standard (720P), Pro (1080P), and 4K tiers, single or multi-shot structure, 3–15s duration, optional audio generation, and up to 2 reference images.
- [Kling 3.0 motion-control API](https://rivya.ai/docs/api/models/kling-3-0-motion-control): Exact 1-image + 1-motion-video project keeps identity and movement roles clear Standard (720P): 20 × verified motion-video seconds; Pro (1080P): 27 × seconds
- [Kling 2.6 API](https://rivya.ai/docs/api/models/kling-2-6): Kuaishou's video model with optional audio generation, 5s/10s clips, and 3 aspect ratios. Strong at human motion and expressive character animation with natural physics.
- [Kling 2.6 motion-control API](https://rivya.ai/docs/api/models/kling-2-6-motion-control): Dedicated motion-transfer project for driving one subject from one reference image plus one motion video. Use it when you want a cheaper Kling motion-control pass and can live without the extra scene controls in Kling 3.0 motion-control.
- [Kling V2.5 Turbo Pro API](https://rivya.ai/docs/api/models/kling-v2-5-turbo-pro): Kuaishou's Kling V2.5 Turbo Pro video model, now supporting both text-to-video and image-to-video. Public pricing evidence clearly covers both text and image tiers at 5 seconds and 10 seconds, so Rivya maps it directly to 42 / 84 credits.
- [Kling V2.1 Master API](https://rivya.ai/docs/api/models/kling-v2-1-master): Kuaishou's older Kling V2.1 Master video model now supports both text-to-video and image-to-video on Rivya. Current pricing is 160 credits for 5 seconds and 320 credits for 10 seconds.
- [Kling V2.1 Pro API](https://rivya.ai/docs/api/models/kling-v2-1-pro): Kuaishou's older Kling V2.1 Pro image-to-video model supports a first frame plus an optional tail frame image. Current pricing is 50 credits for 5 seconds and 100 credits for 10 seconds.
- [Kling V2.1 Standard API](https://rivya.ai/docs/api/models/kling-v2-1-standard): Kuaishou's older Kling V2.1 Standard image-to-video model. Current pricing is 25 credits for 5 seconds and 50 credits for 10 seconds.
- [Kling AI Avatar Pro API](https://rivya.ai/docs/api/models/kling-ai-avatar-pro): Kuaishou's Kling AI Avatar Pro higher-quality talking-avatar model, using one portrait image plus one audio clip to generate lip-synced avatar video. Rivya charges 16 credits per second of verified audio, up to 300 seconds, with the total rounded up to a whole credit.
- [Kling AI Avatar Standard API](https://rivya.ai/docs/api/models/kling-ai-avatar-standard): Kuaishou's Kling AI Avatar Standard talking-avatar model, using one portrait image plus one audio clip to generate lip-synced avatar video. Rivya charges 8 credits per second of verified audio, up to 300 seconds, with the total rounded up to a whole credit.
- [Infinitalk API](https://rivya.ai/docs/api/models/infinitalk): Infinitalk is a portrait-plus-audio talking-video model. Current pricing is metered by resolution and audio duration: 480p = 3 credits per second and 720p = 12 credits per second.
- [Runway API](https://rivya.ai/docs/api/models/runway): Runway is a standalone video model that supports both text-to-video and image-to-video. Public pricing evidence currently confirms only 3 generation tiers, so Rivya keeps it on the verified set: `720p_5 = 12`, `720p_10 = 30`, and `1080p_5 = 30`.
- [Runway Aleph API](https://rivya.ai/docs/api/models/runway-aleph): Source-video transformation project for reworking an existing clip into a new visual result. Use Aleph when the motion comes from your input footage and the creative direction comes from your prompt, with a fixed 90-credit price.
- [Grok Imagine Video 1.5 Preview API](https://rivya.ai/docs/api/models/grok-imagine-video): Grok Imagine Video 1.5 Preview supports 1–15 seconds, 480p or 720p, up to 7 reference images, and prompts up to 4096 characters.
- [Topaz Video Upscaler API](https://rivya.ai/docs/api/models/topaz-video-upscale): One uploaded video, no prompt, with a 1x, 2x, or 4x upscale factor; 2x is the default.
- [Veo3.1 Quality API](https://rivya.ai/docs/api/models/veo3-1-quality): 720p costs 225 credits, 1080p costs 233 (232.5 rounded up), and 4K costs 285; Extend costs 225, while eligible 1080p/4K result actions cost 8/60. Audio is experimental and not guaranteed; review every result manually before delivery. Resolution actions require an owned successful 720p base task or Extend result; a direct 1080p/4K result cannot request another resolution action.
- [Veo3.1 Fast API](https://rivya.ai/docs/api/models/veo3-1-fast): Supports text, first/last-frame, and one-to-three-reference-image generation with direct 720p/1080p/4K and 4/6/8-second controls. Audio is experimental and not guaranteed; review every result manually before delivery.
- [Veo3.1 Lite API](https://rivya.ai/docs/api/models/veo3-1-lite): Supports text, first/last-frame, and one-to-three-reference-image generation with direct 720p/1080p/4K and 4/6/8-second controls. Audio is experimental and not guaranteed; review every result manually before delivery.
- [Wan 3.0 Video API](https://rivya.ai/docs/api/models/wan-3-0-video): Wan 3.0 Video combines Standard and Prime tiers with text-to-video, first/last-frame guidance, and signed multimodal references. It supports 480P, 720P, or 1080P output, explicit 2-30-second clips or intelligent duration, and reference bundles containing images, video, and audio.
- [Kling 3.0 Turbo API](https://rivya.ai/docs/api/models/kling-3-turbo): Kling 3.0 Turbo is Rivya's focused Kling video model for prompt-only generation or animation from one still image. Its two paths share 720p and 1080p output with integer durations from 3 to 15 seconds, while aspect-ratio control stays exclusive to text-to-video.
- [Seedance 2.0 Mini API](https://rivya.ai/docs/api/models/seedance-2-mini): Seedance 2.0 Mini is Rivya's lower-rate multimodal Seedance workflow for prompt-only video, first- or first-and-last-frame animation, and mixed image, video, or audio references. Its 480p and 720p tiers cover 4-to-15-second output while keeping the three input scenes mutually exclusive.
- [Seedance 2.5 API](https://rivya.ai/docs/api/models/seedance-2-5): Seedance 2.5 is Rivya's larger Seedance workflow for prompt-only video, first- and last-frame guidance, multimodal references, and video-guided transformation. It adds a 1080p tier, output up to 30 seconds, automatic duration, MP4 or MOV output, and substantially larger image, video, and audio reference capacity than Mini.
- [MiniMax H3 API](https://rivya.ai/docs/api/models/minimax-h3): MiniMax H3 is Rivya's 768P or 2K video model for prompt-only generation, first- and last-frame animation, and multimodal reference work. Its reference path can combine images, MP4 or MOV clips, and MP3 or WAV audio, while its credit formula separately accounts for verified input-video time and images beyond the first five.
- [HappyHorse 1.1 API](https://rivya.ai/docs/api/models/happyhorse-1-1): HappyHorse 1.1 is a flexible AI video model for prompt-only clips, single-image animation, and multi-image reference video at 720p or 1080p, with 3 to 15 seconds of output.
- [OmniHuman 1.5 API](https://rivya.ai/docs/api/models/omnihuman-1-5): OmniHuman 1.5 turns exactly one portrait image and one audio clip into an audio-driven character video at 720p or 1080p, with credits based on verified audio duration.
- [Volcengine Video Lip Sync API](https://rivya.ai/docs/api/models/volcengine-video-lip-sync): Volcengine Video Lip Sync replaces the spoken performance in one existing video with one target vocal track, using Lite or Basic processing and duration-based credits. The output is MP4 at 25 FPS, and its duration follows the target audio.
- [Kling O3 API](https://rivya.ai/docs/api/models/kling-o3): Verified single-shot generation is available from text, one first frame, or a first-and-last-frame pair.
- [Suno Music API](https://rivya.ai/docs/api/models/suno-music): Suno Music is Rivya's text-to-music model for turning one short brief into a first song draft with or without vocals. It keeps the fixed `12` credit entry point and exposes `Extend Music` as the next step after a successful track.
- [Suno Sounds API](https://rivya.ai/docs/api/models/suno-sounds): Suno Sounds is Rivya's lightweight text-to-sound model for ambience loops, background sound, and short sonic sketches. It keeps the documented fixed price of `2.5` credits per generation and lets successful results continue into `Vocal Separation`.
- [Suno Lyrics API](https://rivya.ai/docs/api/models/suno-lyrics): Suno Lyrics is Rivya's lyric-generation model for turning one theme or mood into song words at a fixed cost of `1` credit per request.
- [ElevenLabs Dialogue V3 API](https://rivya.ai/docs/api/models/elevenlabs-dialogue-v3): ElevenLabs' multi-speaker dialogue model on Rivya. It is built for role-based speech generation, with individual voice assignments, stability controls, and dialogue-ready pacing for podcasts, interviews, and character scenes.
- [ElevenLabs Turbo 2.5 API](https://rivya.ai/docs/api/models/elevenlabs-turbo-2-5): ElevenLabs' fast text-to-speech model on Rivya. With low-latency voice generation and adjustable stability, similarity, style, and speed, it is built for rapid voiceover drafts and interactive TTS projects.
- [ElevenLabs Multilingual V2 API](https://rivya.ai/docs/api/models/elevenlabs-multilingual-v2): ElevenLabs' multilingual text-to-speech model on Rivya, supporting about 30 languages with auto-detection. It is the stronger option for localization, cross-language delivery, and more natural multilingual voiceovers.
- [Claude Opus 4.7 API](https://rivya.ai/docs/api/models/claude-opus-4-7-chat): Anthropic's flagship Claude chat model on Rivya for deep reasoning, careful synthesis, executive writing, and high-impact text work.
- [Claude Sonnet 4.6 API](https://rivya.ai/docs/api/models/claude-sonnet-4-6-chat): Anthropic's balanced Claude AI chat model on Rivya. It keeps strong long-form reasoning and careful analysis for content, research, and coding projects without jumping to Opus-level spend.
- [Claude Sonnet 4.5 API](https://rivya.ai/docs/api/models/claude-sonnet-4-5-chat): Anthropic's balanced Claude AI chat model on Rivya. It is strong at nuanced writing, careful analysis, and safety-conscious responses, making it a strong Claude option for content creation and research.
- [Claude Haiku 4.5 API](https://rivya.ai/docs/api/models/claude-haiku-4-5-chat): Anthropic's lightweight Claude AI chat model on Rivya. It is tuned for speed, cost efficiency, and stable day-to-day chat performance in high-frequency projects where you want Claude tone without premium spend.
- [Claude Opus 4.8 API](https://rivya.ai/docs/api/models/claude-opus-4-8-chat): A text-first Claude chat model on Rivya for reviewing dense source packets and producing careful, structured decisions.
- [Claude Fable 5 API](https://rivya.ai/docs/api/models/claude-fable-5-chat): A text-first Claude chat model on Rivya for narrative reframing, audience-aware drafts, and structured editorial work.
- [Claude Sonnet 5 API](https://rivya.ai/docs/api/models/claude-sonnet-5-chat): A text-first Claude chat model on Rivya for implementation briefs, requirements review, and concise technical planning.
- [Claude Opus 5 API](https://rivya.ai/docs/api/models/claude-opus-5-chat): A text-first Claude chat model on Rivya for board-level risk framing, scenario comparison, and decision documentation.

## Documentation index

- [Rivya AI Help Center](https://rivya.ai/docs): Start with Rivya AI docs for live features, setup, workflows, model paths, tools, credits, billing, uploads, and release checks.
- [Rivya Getting Started Guide](https://rivya.ai/docs/getting-started): Start Rivya with one real task: browse public pages, sign in when needed, understand credits, and choose chat, image, video, or audio.
- [Rivya Glossary](https://rivya.ai/docs/glossary): Understand Rivya terms for Studio, public pages, credits, models, tools, uploads, task states, providers, commercial use, and docs.
- [Rivya Live Features and Product Scope](https://rivya.ai/docs/live-now): Check Rivya's live scope across public pages, models, sign-in requirements, Studio execution, audio coverage, and coming-soon boundaries.
- [Rivya Public Pages vs Signed-In Workflows](https://rivya.ai/docs/public-vs-authenticated-workflows): Understand which Rivya pages are public, which require sign-in, and how prompts, models, uploads, credits, and history move into Studio.
- [Rivya Public Quick Start Guide](https://rivya.ai/docs/public-quick-use): Use Rivya public pages to browse AI models, preview workflows, start quick tasks, compare examples, and decide when to sign in.
- [Rivya Prompt Clipper](https://rivya.ai/docs/rivya-prompt-clipper): Use Rivya Prompt Clipper to turn selected webpage text into local, copy-ready prompts for Rivya image, video, audio, and chat workflows.
- [Rivya Login and Account Access Guide](https://rivya.ai/docs/auth-access): Understand Rivya sign-in with Google or a six-digit email code, deployment-controlled legacy options, recovery paths, protected pages, and account security.
- [Rivya Dashboard Guide](https://rivya.ai/docs/dashboard): Use the Rivya dashboard to check credits, recent generations, chats, notifications, quick actions, account state, and the next workspace.
- [Rivya Studio Guide](https://rivya.ai/docs/studio): Use Rivya Studio for signed-in Chat, Image, Video, and Audio workflows with uploads, credits, history, notifications, and continuity.
- [Rivya Chat Studio Guide](https://rivya.ai/docs/chat-studio): Use Rivya Chat Studio for AI chat, image-assisted analysis, prompt shaping, saved sessions, workflow routing, tools, and creative planning.
- [Rivya Image Studio Guide](https://rivya.ai/docs/image-studio): Use Rivya Image Studio for text-to-image, references, product photos, ecommerce visuals, ad images, saved history, and output review.
- [Rivya Video Studio Guide](https://rivya.ai/docs/video-studio): Use Rivya Video Studio for text-to-video, image-to-video, guided motion, product demos, duration, aspect ratio, audio planning, and review.
- [Rivya Audio Studio Guide](https://rivya.ai/docs/audio-studio): Use Rivya Audio Studio for available voice, narration, dialogue, music, and sound-sketch workflows while checking paused cleanup models.
- [Rivya AI Chat Workflow Guide](https://rivya.ai/docs/chat-workflows): Use Rivya chat for research, prompt shaping, tool sessions, image-assisted analysis, saved context, and image or video handoffs.
- [Rivya AI Image Workflow Guide](https://rivya.ai/docs/image-workflows): Choose Rivya image workflows for product photos, ecommerce stills, references, style exploration, Studio iteration, and saved history.
- [Rivya AI Video Workflow Guide](https://rivya.ai/docs/video-workflows): Choose Rivya video workflows for text-to-video, image-to-video, source video changes, audio-aware clips, aspect ratio, and Studio history.
- [Rivya AI Audio Workflow Guide](https://rivya.ai/docs/audio-workflows): Choose available Rivya audio workflows for voice, dialogue, music, and sound sketches while checking paused sound-effect and cleanup models.
- [Rivya AI Music Workflow Guide](https://rivya.ai/docs/music-workflows): Choose Rivya music workflows for Suno Music drafts, sounds, lyrics, Extend Music, Vocal Separation, WAV conversion, and Studio history.
- [Rivya Content Localization Guide](https://rivya.ai/docs/content-localization): Plan Rivya content localization across text, images, video, voice, captions, examples, and campaign assets while preserving audience meaning.
- [Rivya Prompt Library and Showcase Guide](https://rivya.ai/docs/prompt-library-and-showcase): Use Rivya prompt templates, showcase pages, prompt cards, public starts, Studio handoffs, favorites, source tracking, and return context.
- [Rivya Prompt Writing Guide](https://rivya.ai/docs/prompt-writing-basics): Write Rivya prompts for images, videos, audio, and chat by defining the task, subject, constraints, references, review target, and criteria.
- [Rivya Prompt Templates Guide](https://rivya.ai/docs/prompt-templates): Use Rivya prompt templates for AI image, video, audio, chat, product launches, campaigns, references, constraints, and review checklists.
- [Rivya Model Selection Guide](https://rivya.ai/docs/models): Compare Rivya chat, image, video, and audio models by task, input mode, references, credits, quality settings, and availability.
- [Rivya Model Fields and Parameters Guide](https://rivya.ai/docs/model-fields-and-parameters): Read Rivya model fields for categories, modes, direct generation, credits, references, uploads, quality, duration, aspect ratio, and outputs.
- [Rivya Quality, Duration, and Aspect Ratio Guide](https://rivya.ai/docs/quality-duration-aspect-ratio): Choose Rivya quality, duration, aspect ratio, resolution, cost, speed, and review settings before spending credits on image, video, or audio.
- [Rivya Model Availability Guide](https://rivya.ai/docs/model-availability): Understand Rivya model availability, provider changes, direct generation, workflow support, fallbacks, credit hints, and production readiness.
- [Rivya API Overview](https://rivya.ai/docs/api): Use Rivya API v1 to call Rivya generation and chat models from your own product with API keys, account credits, and optional SSE streaming.
- [Rivya API Quickstart](https://rivya.ai/docs/api-quickstart): Create an API key, choose a model, submit an asynchronous generation job, and send a Chat API turn with optional SSE streaming.
- [API Authentication](https://rivya.ai/docs/api-authentication): Authenticate Rivya API requests with Bearer API keys, scoped permissions, one-time secret display, revocation, and rotation.
- [OpenAPI and Schema Contract](https://rivya.ai/docs/api-openapi-schema): Review Rivya API v1 schema sources, compatibility rules, public fields, and the read-only OpenAPI JSON contract.
- [Rivya TypeScript SDK](https://rivya.ai/docs/api-sdk): Use the Rivya TypeScript SDK beta to call Public API v1 for models, generations, files, credits, webhooks, and Chat including SSE streaming.
- [API Models](https://rivya.ai/docs/api-models): List Rivya API models, understand model IDs, categories, prompt limits, reference media, readiness states, and Files API dependencies.
- [Model API Reference](https://rivya.ai/docs/api-model-reference): Look up Rivya API model IDs, availability, supported modes, parameter tables, prompt limits, reference media rules, and model detail links.
- [Create Generation](https://rivya.ai/docs/api-generations): Submit asynchronous Rivya API generation jobs with model, prompt, params, Idempotency-Key, and public response fields.
- [Chat API](https://rivya.ai/docs/api-chat): Use Rivya Chat API for non-streaming or SSE turns, API-created sessions, file_id image attachments, and token-based credit settlement.
- [Generation Status](https://rivya.ai/docs/api-generation-status): Poll Rivya API generation jobs by public task ID, read queued, processing, succeeded, and failed states, and consume result URLs.
- [API Credits](https://rivya.ai/docs/api-credits): Understand how Rivya API calls use account credits, balance checks, reserved credits, failed task refunds, and credit troubleshooting.
- [API Errors and Limits](https://rivya.ai/docs/api-errors-and-limits): Handle Rivya API public error codes, HTTP status values, rate limits, idempotency conflicts, and retry decisions.
- [Files API](https://rivya.ai/docs/api-files): Upload image, video, or audio reference files for Rivya API generation requests, with MIME checks, size limits, and duration tokens.
- [API Webhooks](https://rivya.ai/docs/api-webhooks): Create signed Rivya API webhook endpoints, verify delivery signatures, inspect delivery attempts, and send safe test events.
- [API Changelog](https://rivya.ai/docs/api-changelog): Track Rivya API v1 documentation, endpoint, model reference, schema, and future surface updates.
- [Rivya References and Uploads Guide](https://rivya.ai/docs/references-and-uploads): Plan Rivya references, audio uploads, file limits, sign-in requirements, safety checks, model choice, and Studio execution.
- [Rivya Image References Guide](https://rivya.ai/docs/image-references): Use Rivya image references to guide product identity, composition, style, lighting, brand consistency, generation, and output review.
- [Rivya Video References Guide](https://rivya.ai/docs/video-references): Use Rivya video references to guide product appearance, scene structure, motion, camera behavior, image-to-video continuity, and review.
- [Rivya Audio Uploads Guide](https://rivya.ai/docs/audio-uploads): Prepare Rivya audio uploads safely, check model availability, and understand why the documented cleanup and isolation path is currently paused.
- [Rivya Safe Upload Guidelines](https://rivya.ai/docs/safe-upload-guidelines): Review privacy, rights, sensitive data, people, voice, brand assets, provider processing, and file necessity before uploading to Rivya.
- [Rivya Browser and File Requirements](https://rivya.ai/docs/browser-and-file-requirements): Check Rivya browser, network, upload, file type, size, output, and retry requirements before image, video, audio, or chat workflows.
- [Rivya AI Tools Guide](https://rivya.ai/docs/tools): See Rivya AI tools, how AI Calculator and AI Solver connect to chat, when to use tools, and how to read coming-soon categories.
- [Rivya Credits and Billing Guide](https://rivya.ai/docs/credits-billing): Understand Rivya credit rules, signup credits, plans, packs, FIFO spending, expiry, Stripe Checkout, low-balance behavior, and wallet checks.
- [Rivya Plans and Credit Packs Guide](https://rivya.ai/docs/plans-and-packs): Compare Rivya Free, Basic, Advanced, Pro, Premium, and credit packs for recurring AI work, project bursts, and one-time top-ups.
- [Rivya Pricing FAQ](https://rivya.ai/docs/pricing-faq): Get answers about Rivya Free, signup credits, plans, credit packs, failed runs, Checkout, refunds, cancellations, and billing management.
- [Rivya Payment Checkout Guide](https://rivya.ai/docs/payment-checkout): Understand Rivya plan and credit-pack checkout, Stripe redirects, the /payment bridge, webhooks, billing updates, and purchase checks.
- [Rivya Commercial Use and Ownership Guide](https://rivya.ai/docs/commercial-use-and-rights): Review ownership, input rights, provider terms, claims, approvals, and client delivery before using Rivya AI outputs commercially.
- [Rivya Data and Provider Processing Guide](https://rivya.ai/docs/data-and-provider-processing): Understand what Rivya stores, when providers may process prompts, uploads, outputs, and metadata, and how to handle sensitive data.
- [Rivya Provider and Commercial-Use Matrix](https://rivya.ai/docs/provider-and-commercial-use-matrix): Review Rivya workflows by provider touchpoints, ownership framing, data sensitivity, upload rights, and pre-publication commercial checks.
- [Rivya Commercial Review Checklist](https://rivya.ai/docs/commercial-review-checklist): Review Rivya AI assets before publishing or client delivery: accuracy, claims, rights, uploads, logos, likeness, brand fit, and approval.
- [Rivya Task Lifecycle Guide](https://rivya.ai/docs/task-lifecycle): Understand Rivya task status, credit reservation, provider submission, callbacks, polling, history, notifications, failures, and credits.
- [Rivya Failed Tasks and Credit Refunds Guide](https://rivya.ai/docs/failed-tasks-and-refunds): Handle Rivya failed tasks, credit checks, retries, upload issues, provider errors, notifications, history, and processing states.
- [Generation Queue and Waiting Times](https://rivya.ai/docs/generation-queue-and-waiting): Understand Rivya generation queues, waiting times, long-running image, video, and audio tasks, notifications, history, and safe retries.
- [Rivya History Guide](https://rivya.ai/docs/history): Use Rivya history to reopen chats, generation results, images, videos, audio outputs, task status, downloads, and project context.
- [Rivya Saved Sessions Guide](https://rivya.ai/docs/saved-sessions): Use Rivya saved sessions to continue chats, prompt drafts, tool conversations, generation context, project decisions, and longer workflows.
- [Rivya Output Downloads and Reuse Guide](https://rivya.ai/docs/output-downloads): Review, download, reuse, and hand off Rivya AI outputs across images, videos, audio, history, commercial checks, and brand review.
- [Rivya Notifications Center Guide](https://rivya.ai/docs/notifications-center): Use Rivya notifications to track billing events, credits, generation results, low balance, dashboard previews, unread status, and follow-ups.
- [Rivya Account Settings Guide](https://rivya.ai/docs/account-settings): Manage Rivya profile, billing, credits, security, avatar, newsletters, password options, account deletion, and authentication settings.
- [Rivya Troubleshooting Guide](https://rivya.ai/docs/troubleshooting): Fix Rivya issues with chat sending, uploads, stuck generation tasks, missing results, payment updates, credits, history, and notifications.