# Core.Today AI API

> Unified API gateway for multiple AI providers with credit-based usage tracking.

Base URL: https://api.core.today/v1
Documentation: https://console.core.today/docs

## Integration guides (start here to build an integration)

- [API Integration Guide](https://console.core.today/llms/api.txt): Full service guide — auth, model discovery & pricing, predictions flow, file uploads, LLM gateway, databases, templates, credits, rate limits, errors.
- [Reseller / End-Customer Budgets](https://console.core.today/llms/reseller.txt): Resell AI capacity to your own end-customers — customer wallets, topups, monthly budgets, X-Customer-Id attribution, phantom tokens, invoice-grade usage exports.

## Category guides (full model specs per category)

- [Image Generation](https://console.core.today/llms/image-models.txt)
- [Video Generation](https://console.core.today/llms/video-models.txt)
- [Audio & TTS](https://console.core.today/llms/audio-models.txt)
- [Music Generation](https://console.core.today/llms/music-generation.txt)
- [LLM Chat Completions](https://console.core.today/llms/llm.txt)

## Image Generation

- **[Background Remover](https://console.core.today/llms/models/851-labs/background-remover.txt)**: 851-labs/background-remover | image | 1 credits — The most-used background remover on Replicate (27M+ runs).
- **[BiRefNet](https://console.core.today/llms/models/men1scus/birefnet.txt)**: men1scus/birefnet | image | 5 credits — State-of-the-art open-source background removal.
- **[BLIP](https://console.core.today/llms/models/salesforce/blip.txt)**: salesforce/blip | image | 1 credits — Salesforce BLIP (173M+ runs) — image captioning, visual question answering, and image-text matching in one model.
- **[Change Haircut](https://console.core.today/llms/models/flux-kontext-apps/change-haircut.txt)**: flux-kontext-apps/change-haircut | image | 93 credits — Change anyone's hairstyle and hair color from a single photo, powered by FLUX.
- **[Clarity Upscaler](https://console.core.today/llms/models/philz1337x/clarity-upscaler.txt)**: philz1337x/clarity-upscaler | image | 60 credits — The famous creative upscaler (30M+ runs).
- **[CLIP Features](https://console.core.today/llms/models/andreasjansson/clip-features.txt)**: andreasjansson/clip-features | image | 1 credits — CLIP ViT-L/14 embeddings for text AND images (163M+ runs) — puts both in the same vector space for cross-modal search, image dedup, and zero-shot classification.
- **[CodeFormer](https://console.core.today/llms/models/sczhou/codeformer.txt)**: sczhou/codeformer | image | 8 credits — Robust face restoration for old photos and AI-generated faces (54M+ runs).
- **[ControlNet Scribble](https://console.core.today/llms/models/jagilley/controlnet-scribble.txt)**: jagilley/controlnet-scribble | image | 120 credits — The classic sketch-to-image model (38M+ runs) — turn any scribble or line drawing into a detailed image guided by your prompt.
- **[Nano Banana (Edit)](https://console.core.today/llms/models/fal-ai/nano-banana/edit.txt)**: fal-ai/nano-banana/edit | image | 91 credits — Dedicated edit endpoint for Nano Banana, Google's Gemini 2.
- **[Nano Banana 2 (Edit)](https://console.core.today/llms/models/fal-ai/nano-banana-2/edit.txt)**: fal-ai/nano-banana-2/edit | image | 190 credits — Edit endpoint for Nano Banana 2, built on Gemini 3.
- **[Nano Banana Pro (Edit)](https://console.core.today/llms/models/fal-ai/nano-banana-pro/edit.txt)**: fal-ai/nano-banana-pro/edit | image | 350 credits — Edit endpoint for Nano Banana Pro built on Gemini 3 Pro.
- **[Face to Many](https://console.core.today/llms/models/fofr/face-to-many.txt)**: fofr/face-to-many | image | 21 credits — Turn a face photo into 6 fun styles (15M+ runs) — 3D, Emoji, Video game, Pixels, Clay, or Toy.
- **[Face to Sticker](https://console.core.today/llms/models/fofr/face-to-sticker.txt)**: fofr/face-to-sticker | image | 22 credits — Turn any face photo into a fun die-cut sticker (1.
- **[Florence-2 Large](https://console.core.today/llms/models/lucataco/florence-2-large.txt)**: lucataco/florence-2-large | image | 2 credits — Microsoft's Florence-2 all-in-one vision model — captioning, object detection, phrase grounding, OCR, and segmentation in a single API.
- **[FLUX.2 Klein 4B](https://console.core.today/llms/models/black-forest-labs/flux-2-klein-4b.txt)**: black-forest-labs/flux-2-klein-4b | image | 2 credits — Very fast image generation and editing model.
- **[FLUX 2 Flex](https://console.core.today/llms/models/black-forest-labs/flux-2-flex.txt)**: black-forest-labs/flux-2-flex | image | 280 credits — Maximum-quality FLUX model supporting up to 10 reference images and advanced typography.
- **[FLUX 2 Max](https://console.core.today/llms/models/black-forest-labs/flux-2-max.txt)**: black-forest-labs/flux-2-max | image | 160 credits — The highest fidelity image model from Black Forest Labs.
- **[FLUX 2 Pro](https://console.core.today/llms/models/black-forest-labs/flux-2-pro.txt)**: black-forest-labs/flux-2-pro | image | 70 credits — Professional-grade FLUX 2 with high-quality editing and up to 8 reference image support.
- **[FLUX.2 Dev](https://console.core.today/llms/models/black-forest-labs/flux-2-dev.txt)**: black-forest-labs/flux-2-dev | image | 28 credits — Development version of FLUX.
- **[FLUX 1.1 Pro](https://console.core.today/llms/models/black-forest-labs/flux-1.1-pro.txt)**: black-forest-labs/flux-1.1-pro | image | 93 credits — Fast high-quality image generation, an upgrade to FLUX.
- **[FLUX 1.1 Pro Ultra](https://console.core.today/llms/models/black-forest-labs/flux-1.1-pro-ultra.txt)**: black-forest-labs/flux-1.1-pro-ultra | image | 140 credits — FLUX1.
- **[FLUX Dev](https://console.core.today/llms/models/black-forest-labs/flux-dev.txt)**: black-forest-labs/flux-dev | image | 58 credits — A 12 billion parameter rectified flow transformer capable of generating images from text descriptions, tuned for open, high-quality experimentation.
- **[FLUX Kontext Max](https://console.core.today/llms/models/black-forest-labs/flux-kontext-max.txt)**: black-forest-labs/flux-kontext-max | image | 190 credits — A premium text-based image editing model that delivers maximum performance and improved typography generation for transforming images through natural language prompts.
- **[FLUX Kontext Pro](https://console.core.today/llms/models/black-forest-labs/flux-kontext-pro.txt)**: black-forest-labs/flux-kontext-pro | image | 93 credits — State-of-the-art text-based image editing model that transforms images through natural language.
- **[FLUX.1 Krea [dev]](https://console.core.today/llms/models/black-forest-labs/flux-krea-dev.txt)**: black-forest-labs/flux-krea-dev | image | 58 credits — Photorealistic image generation that specifically avoids the 'AI look', producing natural-looking images indistinguishable from real photographs.
- **[FLUX PuLID](https://console.core.today/llms/models/bytedance/flux-pulid.txt)**: bytedance/flux-pulid | image | 47 credits — PuLID identity customization on FLUX-dev: generate photorealistic portraits of a specific person from one face photo, with markedly higher fidelity than SDXL-based variants.
- **[FLUX.1 Schnell](https://console.core.today/llms/models/black-forest-labs/flux-schnell.txt)**: black-forest-labs/flux-schnell | image | 7 credits — Ultra-fast image generation model optimized for speed.
- **[Runway Gen-4 Image](https://console.core.today/llms/models/runwayml/gen4-image.txt)**: runwayml/gen4-image | image | 190 credits — Runway's Gen-4 Image model with references: combine up to 3 reference images with @tag mentions in your prompt to keep characters, objects, and locations consistent across every angle and scene, at 720p or 1080p.
- **[Runway Gen-4 Image Turbo](https://console.core.today/llms/models/runwayml/gen4-image-turbo.txt)**: runwayml/gen4-image-turbo | image | 70 credits — Gen-4 Image Turbo is 2.
- **[GFPGAN](https://console.core.today/llms/models/tencentarc/gfpgan.txt)**: tencentarc/gfpgan | image | 6 credits — Tencent ARC's face restoration model — 115M+ runs, the most-run model in the restoration collection.
- **[GPT Image 2](https://console.core.today/llms/models/openai/gpt-image-2.txt)**: openai/gpt-image-2 | image | 300 credits — OpenAI's state-of-the-art image generation and editing model with strong instruction following, sharp text rendering, and detailed editing.
- **[GPT Image 1.5](https://console.core.today/llms/models/openai/gpt-image-1.5.txt)**: openai/gpt-image-1.5 | image | 310 credits — OpenAI's latest image generation model with better instruction following and adherence to prompts, including sharp text rendering and detailed editing.
- **[Grok Imagine Image](https://console.core.today/llms/models/xai/grok-imagine-image.txt)**: xai/grok-imagine-image | image | 47 credits — Generate images using xAI's Grok Imagine model.
- **[Grounding DINO](https://console.core.today/llms/models/adirik/grounding-dino.txt)**: adirik/grounding-dino | image | 2 credits — Text-prompted object detection (39M+ runs) — describe what to find in natural language ('red car, person wearing a hat') and get bounding boxes with confidence scores plus an annotated image.
- **[Hunyuan 3D 3.1](https://console.core.today/llms/models/tencent/hunyuan-3d-3.1.txt)**: tencent/hunyuan-3d-3.1 | image | 1160 credits — Tencent's flagship 3D generation — create high-polygon textured 3D models from a text prompt OR an image, with optional PBR (physically based rendering) materials.
- **[Ideogram V4 Balanced](https://console.core.today/llms/models/ideogram-ai/ideogram-v4-balanced.txt)**: ideogram-ai/ideogram-v4-balanced | image | 140 credits — A middle-ground tier in Ideogram's v4 family, balancing generation speed and output quality.
- **[Ideogram V4 Quality](https://console.core.today/llms/models/ideogram-ai/ideogram-v4-quality.txt)**: ideogram-ai/ideogram-v4-quality | image | 230 credits — The highest-fidelity tier of Ideogram's v4 model family, tuned for maximum detail, realism, and typography accuracy.
- **[Ideogram V4 Turbo](https://console.core.today/llms/models/ideogram-ai/ideogram-v4-turbo.txt)**: ideogram-ai/ideogram-v4-turbo | image | 70 credits — The fastest and cheapest model in Ideogram's v4 family, built for rapid iteration while retaining Ideogram's signature text rendering and style consistency.
- **[Ideogram V3 Turbo](https://console.core.today/llms/models/ideogram-ai/ideogram-v3-turbo.txt)**: ideogram-ai/ideogram-v3-turbo | image | 70 credits — The fastest and cheapest Ideogram v3 tier.
- **[Ideogram Character](https://console.core.today/llms/models/ideogram-ai/ideogram-character.txt)**: ideogram-ai/ideogram-character | image | 350 credits — Generate consistent characters from a single reference image.
- **[MiniMax Image-01](https://console.core.today/llms/models/minimax/image-01.txt)**: minimax/image-01 | image | 23 credits — MiniMax's first image generation model with character reference support: provide a single face photo via subject_reference and generate consistent images of that person across prompts, styles, and aspect ratios — up to 9 images per request.
- **[Topaz Image Upscale](https://console.core.today/llms/models/topazlabs/image-upscale.txt)**: topazlabs/image-upscale | image | 190 credits — Professional-grade upscaling from Topaz Labs, the industry standard for photo enhancement.
- **[Imagen 4](https://console.core.today/llms/models/google/imagen-4.txt)**: google/imagen-4 | image | 93 credits — Google's Imagen 4 flagship text-to-image model.
- **[Imagen 4 Fast](https://console.core.today/llms/models/google/imagen-4-fast.txt)**: google/imagen-4-fast | image | 46 credits — A fast version of Imagen 4 for when speed and cost are more important than maximum quality.
- **[Bria Increase Resolution](https://console.core.today/llms/models/bria/increase-resolution.txt)**: bria/increase-resolution | image | 93 credits — Bria's commercially-safe image upscaler (130K+ runs).
- **[Krea 2 Large](https://console.core.today/llms/models/krea/krea-2-large.txt)**: krea/krea-2-large | image | 140 credits — Krea AI's flagship text-to-image model, focused on photorealistic output with strong prompt adherence.
- **[Krea 2 Medium](https://console.core.today/llms/models/krea/krea-2-medium.txt)**: krea/krea-2-medium | image | 70 credits — A lower-cost variant of Krea 2 Large, trading some fidelity for faster and cheaper generation while keeping the same photorealistic style focus.
- **[Microsoft MAI-Image 2.5 Pro](https://console.core.today/llms/models/microsoft/mai-image-2.5-pro.txt)**: microsoft/mai-image-2.5-pro | image | 400 credits — Microsoft's highest-fidelity image model for production-grade text-to-image generation via Fal.
- **[Moondream2](https://console.core.today/llms/models/lucataco/moondream2.txt)**: lucataco/moondream2 | image | 5 credits — Small but capable vision-language model (14M+ runs) — ask free-form questions about any image and get detailed answers.
- **[Multilingual E5 Large](https://console.core.today/llms/models/beautyyuyanli/multilingual-e5-large.txt)**: beautyyuyanli/multilingual-e5-large | image | 2 credits — Multilingual text embeddings (74M+ runs) — 1024-dimension vectors across 100 languages including Korean.
- **[Nano Banana 2 (Gemini 3.1 Flash Image)](https://console.core.today/llms/models/google/nano-banana-2.txt)**: google/nano-banana-2 | image | 190 credits — Google's fast image generation model built on Gemini 3.
- **[Nano Banana 2 Lite](https://console.core.today/llms/models/google/nano-banana-2-lite.txt)**: google/nano-banana-2-lite | image | 80 credits — Google's lightweight Nano Banana 2 variant built on Gemini 3.
- **[Nano Banana](https://console.core.today/llms/models/google/nano-banana.txt)**: google/nano-banana | image | 91 credits — Google Gemini 2.
- **[Nano Banana Pro (Gemini 3 Pro Image)](https://console.core.today/llms/models/google/nano-banana-pro.txt)**: google/nano-banana-pro | image | 350 credits — Google's state-of-the-art image generation and editing model built on Gemini 3 Pro.
- **[NSFW Image Detection](https://console.core.today/llms/models/falcons-ai/nsfw_image_detection.txt)**: falcons-ai/nsfw_image_detection | image | 1 credits — The standard NSFW image classifier (127M+ runs) — returns 'normal' or 'nsfw' for any image.
- **[Pruna P-Image](https://console.core.today/llms/models/prunaai/p-image.txt)**: prunaai/p-image | image | 12 credits — Pruna AI's distilled text-to-image model optimized for extremely low-cost, high-throughput generation.
- **[Pruna P-Image Upscale](https://console.core.today/llms/models/prunaai/p-image-upscale.txt)**: prunaai/p-image-upscale | image | 12 credits — PrunaAI's high-end image upscaler (575K+ runs) reaching up to 128-megapixel output.
- **[PhotoMaker](https://console.core.today/llms/models/tencentarc/photomaker.txt)**: tencentarc/photomaker | image | 10 credits — Generate stylized photos of a person from 1-4 reference photos (9M+ runs).
- **[PhotoMaker Style](https://console.core.today/llms/models/tencentarc/photomaker-style.txt)**: tencentarc/photomaker-style | image | 15 credits — The stylization-focused variant of PhotoMaker: turn 1-4 photos of a person into paintings, comics, 3D art, and more with stronger style transfer.
- **[Professional Headshot](https://console.core.today/llms/models/flux-kontext-apps/professional-headshot.txt)**: flux-kontext-apps/professional-headshot | image | 93 credits — Turn any single photo into a polished professional business headshot, powered by FLUX.
- **[Proteus v0.2](https://console.core.today/llms/models/datacte/proteus-v0.2.txt)**: datacte/proteus-v0.2 | image | 35 credits — Popular anime-focused image model (12M+ runs) — high-quality anime and illustration styles with img2img and inpainting support.
- **[PuLID](https://console.core.today/llms/models/bytedance/pulid.txt)**: bytedance/pulid | image | 6 credits — ByteDance PuLID: tuning-free identity customization on SDXL.
- **[Real-ESRGAN](https://console.core.today/llms/models/nightmareai/real-esrgan.txt)**: nightmareai/real-esrgan | image | 5 credits — The classic image upscaler (94M+ runs).
- **[Recraft V4](https://console.core.today/llms/models/recraft-ai/recraft-v4.txt)**: recraft-ai/recraft-v4 | image | 93 credits — Recraft's next-generation text-to-image model, improving on V3's typography, style range, and prompt adherence for production-grade brand and design assets.
- **[Recraft V4 SVG](https://console.core.today/llms/models/recraft-ai/recraft-v4-svg.txt)**: recraft-ai/recraft-v4-svg | image | 190 credits — Generates vector graphics directly in SVG format using Recraft V4, ideal for logos, icons, and scalable illustrations that need to stay crisp at any size.
- **[Recraft V3](https://console.core.today/llms/models/recraft-ai/recraft-v3.txt)**: recraft-ai/recraft-v3 | image | 93 credits — Recraft V3 (code-named red_panda) is a text-to-image model with the ability to generate long texts, and images in a wide list of styles.
- **[Recraft Crisp Upscale](https://console.core.today/llms/models/recraft-ai/recraft-crisp-upscale.txt)**: recraft-ai/recraft-crisp-upscale | image | 14 credits — Fast, affordable upscaler from Recraft designed for sharp, crisp results — a single image input with no tuning needed.
- **[Remove Background](https://console.core.today/llms/models/bria/remove-background.txt)**: bria/remove-background | image | 32 credits — AI-powered background removal tool for images.
- **[Remove BG](https://console.core.today/llms/models/lucataco/remove-bg.txt)**: lucataco/remove-bg | image | 1 credits — One of the most-run background removers on Replicate (17M+ runs).
- **[Stable Diffusion XL](https://console.core.today/llms/models/stability-ai/sdxl.txt)**: stability-ai/sdxl | image | 9 credits — Stability AI's classic SDXL (85M+ runs) — the battle-tested text-to-image model with img2img, inpainting, refiner, and LoRA support.
- **[SDXL Lightning 4-step](https://console.core.today/llms/models/bytedance/sdxl-lightning-4step.txt)**: bytedance/sdxl-lightning-4step | image | 4 credits — ByteDance's 4-step SDXL Lightning — the most-run model on Replicate (1B+ runs).
- **[Seedream 5 Lite](https://console.core.today/llms/models/bytedance/seedream-5-lite.txt)**: bytedance/seedream-5-lite | image | 82 credits — Seedream 5.
- **[Seedream 5 Pro](https://console.core.today/llms/models/bytedance/seedream-5-pro.txt)**: bytedance/seedream-5-pro | image | 210 credits — ByteDance's flagship Seedream 5.
- **[Seedream 4.5](https://console.core.today/llms/models/bytedance/seedream-4.5.txt)**: bytedance/seedream-4.5 | image | 90 credits — Upgraded ByteDance image model with stronger spatial understanding and world knowledge.
- **[Seedream 4.0](https://console.core.today/llms/models/bytedance/seedream-4.txt)**: bytedance/seedream-4 | image | 70 credits — ByteDance's latest image generation model with exceptional prompt understanding and creative capabilities.
- **[Text Extract OCR](https://console.core.today/llms/models/abiruyt/text-extract-ocr.txt)**: abiruyt/text-extract-ocr | image | 1 credits — Simple, massively-used OCR (91M+ runs) — extracts text from an image with a single input and returns plain text.
- **[TRELLIS](https://console.core.today/llms/models/firtoz/trellis.txt)**: firtoz/trellis | image | 77 credits — Microsoft's TRELLIS image-to-3D (836K+ runs — the most-used 3D model on Replicate).
- **[Google Upscaler](https://console.core.today/llms/models/google/upscaler.txt)**: google/upscaler | image | 46 credits — Google's image upscaler (770K+ runs).
- **[Z-Image Turbo](https://console.core.today/llms/models/prunaai/z-image-turbo.txt)**: prunaai/z-image-turbo | image | 40 credits — Pruna's ultra-fast Z-Image Turbo (48M+ runs).

## Video Generation

- **[Add Watermark](https://console.core.today/llms/models/charlesmccarthy/addwatermark.txt)**: charlesmccarthy/addwatermark | video | 1 credits — Add a text watermark to any video — simple, fast brand protection for generated or user content at 2 credits per video.
- **[CogVLM2 Video](https://console.core.today/llms/models/chenxwh/cogvlm2-video.txt)**: chenxwh/cogvlm2-video | video | 28 credits — Video understanding and captioning — ask free-form questions about a video and get detailed answers about actions, scenes, and content.
- **[Google Gemini Omni Flash](https://console.core.today/llms/models/google/gemini-omni-flash.txt)**: google/gemini-omni-flash | video | 2320 credits — Google Gemini Omni Flash text-to-video via Fal.
- **[Gen-4.5](https://console.core.today/llms/models/runwayml/gen-4.5.txt)**: runwayml/gen-4.5 | video | 1400 credits — Runway's Gen-4.
- **[Grok Imagine Video 1.5](https://console.core.today/llms/models/xai/grok-imagine-video-1.5.txt)**: xai/grok-imagine-video-1.5 | video | 930 credits — xAI's Grok Imagine Video 1.
- **[Grok Imagine Video](https://console.core.today/llms/models/xai/grok-imagine-video.txt)**: xai/grok-imagine-video | video | 580 credits — Generate videos using xAI's Grok Imagine Video model.
- **[MiniMax Hailuo 2.3](https://console.core.today/llms/models/minimax/hailuo-2.3.txt)**: minimax/hailuo-2.3 | video | 650 credits — Realistic human motion video generation with advanced character consistency and natural movement.
- **[MiniMax Hailuo 2.3 Fast](https://console.core.today/llms/models/minimax/hailuo-2.3-fast.txt)**: minimax/hailuo-2.3-fast | video | 450 credits — Lower-latency version of Hailuo 2.
- **[Happy Horse 1.1](https://console.core.today/llms/models/alibaba/happyhorse-1.1.txt)**: alibaba/happyhorse-1.1 | video | 2090 credits — Alibaba's Happy Horse 1.
- **[MiniMax Hailuo-03 (H3) Image to Video](https://console.core.today/llms/models/minimax/h3/image-to-video.txt)**: minimax/h3/image-to-video | video | 3020 credits — MiniMax Hailuo-03 (H3) image-to-video via Fal.
- **[Kling v3 Omni Video](https://console.core.today/llms/models/kwaivgi/kling-v3-omni-video.txt)**: kwaivgi/kling-v3-omni-video | video | 2610 credits — Kling Video 3.
- **[Kling v3 Video](https://console.core.today/llms/models/kwaivgi/kling-v3-video.txt)**: kwaivgi/kling-v3-video | video | 2610 credits — Kling Video 3.
- **[Kling v2.6](https://console.core.today/llms/models/kwaivgi/kling-v2.6.txt)**: kwaivgi/kling-v2.6 | video | 1630 credits — Kling 2.
- **[Kling 2.5 Turbo Pro](https://console.core.today/llms/models/kwaivgi/kling-v2.5-turbo-pro.txt)**: kwaivgi/kling-v2.5-turbo-pro | video | 810 credits — Cinematic-grade video generation with enhanced motion and scene coherence.
- **[Kling v2.1](https://console.core.today/llms/models/kwaivgi/kling-v2.1.txt)**: kwaivgi/kling-v2.1 | video | 580 credits — Kling v2.
- **[LatentSync](https://console.core.today/llms/models/bytedance/latentsync.txt)**: bytedance/latentsync | video | 220 credits — ByteDance's open-source lipsync — re-syncs a video's mouth movements to any audio track using latent diffusion.
- **[MMAudio](https://console.core.today/llms/models/zsxkib/mmaudio.txt)**: zsxkib/mmaudio | video | 11 credits — Add AI-generated sound to any video (5M+ runs) — synthesizes synchronized audio (ambience, effects, foley) from the video content and an optional text prompt.
- **[Pruna P-Video](https://console.core.today/llms/models/prunaai/p-video.txt)**: prunaai/p-video | video | 240 credits — PrunaAI's fast video generator with a built-in draft mode for rapid creative iteration.
- **[Pruna P-Video Avatar](https://console.core.today/llms/models/prunaai/p-video-avatar.txt)**: prunaai/p-video-avatar | video | 870 credits — Pruna AI's talking-avatar video model — animates a single input image to speak provided text (voice_script) or an uploaded audio track, with selectable Gemini-family preset voices, language, and visual delivery prompt.
- **[PixVerse V6](https://console.core.today/llms/models/pixverse/pixverse-v6.txt)**: pixverse/pixverse-v6 | video | 810 credits — PixVerse's flagship video generation model.
- **[PixVerse V5](https://console.core.today/llms/models/pixverse/pixverse-v5.txt)**: pixverse/pixverse-v5 | video | 24 credits — Advanced video generation with special effects capabilities and anime-optimized output, supporting multiple visual styles.
- **[Luma Ray 3.2](https://console.core.today/llms/models/luma/ray-3.2.txt)**: luma/ray-3.2 | video | 1740 credits — Luma's flagship Ray video model.
- **[Luma Ray Flash 2 720p](https://console.core.today/llms/models/luma/ray-flash-2-720p.txt)**: luma/ray-flash-2-720p | video | 700 credits — Luma's Ray Flash 2 generates 5 or 9 second 720p videos faster and cheaper than Ray 2.
- **[Real-ESRGAN Video](https://console.core.today/llms/models/lucataco/real-esrgan-video.txt)**: lucataco/real-esrgan-video | video | 1280 credits — Video upscaling with Real-ESRGAN — enhance videos to FHD, 2K, or 4K frame by frame.
- **[SadTalker](https://console.core.today/llms/models/cjwbw/sadtalker.txt)**: cjwbw/sadtalker | video | 91 credits — Talking-head video from a single photo and an audio track — animates the face with natural head motion and eye blinks, with optional GFPGAN face enhancement.
- **[Seedance 2.0](https://console.core.today/llms/models/bytedance/seedance-2.0.txt)**: bytedance/seedance-2.0 | video | 2090 credits — ByteDance's next-generation multimodal video model with native synchronized audio.
- **[Seedance 2.0 Fast](https://console.core.today/llms/models/bytedance/seedance-2.0-fast.txt)**: bytedance/seedance-2.0-fast | video | 1750 credits — A faster, cheaper variant of Seedance 2.
- **[Seedance 2.0 Mini](https://console.core.today/llms/models/bytedance/seedance-2.0-mini.txt)**: bytedance/seedance-2.0-mini | video | 1280 credits — Lighter, cheaper variant of ByteDance's Seedance 2.
- **[Seedance 1 Lite](https://console.core.today/llms/models/bytedance/seedance-1-lite.txt)**: bytedance/seedance-1-lite | video | 420 credits — A lightweight ByteDance video generation model offering text-to-video and image-to-video support for 4-12 second videos at 480p, 720p, or 1080p resolution.
- **[Seedance 1 Pro](https://console.core.today/llms/models/bytedance/seedance-1-pro.txt)**: bytedance/seedance-1-pro | video | 700 credits — A pro version of Seedance that offers text-to-video and image-to-video support for 2-12 second videos, at 480p, 720p, and 1080p resolution.
- **[Seedance 1 Pro Fast](https://console.core.today/llms/models/bytedance/seedance-1-pro-fast.txt)**: bytedance/seedance-1-pro-fast | video | 290 credits — ByteDance's cinematic video generation model with fast generation speed and professional output quality.
- **[OpenAI Sora 2](https://console.core.today/llms/models/openai/sora-2.txt)**: openai/sora-2 | video | 930 credits — OpenAI's video generation model with realistic physics simulation and audio generation capabilities, producing highly coherent videos.
- **[OpenAI Sora 2 Pro](https://console.core.today/llms/models/openai/sora-2-pro.txt)**: openai/sora-2-pro | video | 2790 credits — OpenAI's most advanced synced-audio video generation model.
- **[MiniMax Hailuo-03 (H3) Text to Video](https://console.core.today/llms/models/minimax/h3/text-to-video.txt)**: minimax/h3/text-to-video | video | 3020 credits — MiniMax Hailuo-03 (H3) text-to-video via Fal.
- **[Veed Lipsync v2](https://console.core.today/llms/models/veed/lipsync/v2.txt)**: veed/lipsync/v2 | video | 2440 credits — Veed Lipsync v2 via Fal.
- **[Google Veo 3.1](https://console.core.today/llms/models/google/veo-3.1.txt)**: google/veo-3.1 | video | 7440 credits — Google's state-of-the-art video generation model with built-in audio generation, producing cinematic-quality videos with synchronized sound.
- **[Google Veo 3.1 Fast](https://console.core.today/llms/models/google/veo-3.1-fast.txt)**: google/veo-3.1-fast | video | 2790 credits — Fast version of Veo 3.
- **[Video Utils](https://console.core.today/llms/models/nicolascoutureau/video-utils.txt)**: nicolascoutureau/video-utils | video | 5 credits — FFmpeg-powered video utilities (20M+ runs) — convert to mp4/gif, extract audio as mp3, or dump zipped frames, all with one task parameter.
- **[Wan 2.7 I2V](https://console.core.today/llms/models/wan-video/wan-2.7-i2v.txt)**: wan-video/wan-2.7-i2v | video | 1740 credits — Alibaba Wan 2.
- **[Wan 2.7 T2V](https://console.core.today/llms/models/wan-video/wan-2.7-t2v.txt)**: wan-video/wan-2.7-t2v | video | 1740 credits — Alibaba Wan 2.
- **[Wan 2.5 I2V](https://console.core.today/llms/models/wan-video/wan-2.5-i2v.txt)**: wan-video/wan-2.5-i2v | video | 1160 credits — Image-to-video model with lip sync support, animating still images into realistic videos with natural motion.
- **[Wan 2.5 I2V Fast](https://console.core.today/llms/models/wan-video/wan-2.5-i2v-fast.txt)**: wan-video/wan-2.5-i2v-fast | video | 790 credits — Fast image-to-video variant of Wan 2.
- **[Wan 2.5 T2V](https://console.core.today/llms/models/wan-video/wan-2.5-t2v.txt)**: wan-video/wan-2.5-t2v | video | 2320 credits — Text-to-video model with audio synchronization support, producing high-quality videos from text prompts with natural motion.
- **[Wan 2.5 T2V Fast](https://console.core.today/llms/models/wan-video/wan-2.5-t2v-fast.txt)**: wan-video/wan-2.5-t2v-fast | video | 790 credits — Fast text-to-video generation variant of Wan 2.
- **[Wan 2.2 I2V Fast](https://console.core.today/llms/models/wan-video/wan-2.2-i2v-fast.txt)**: wan-video/wan-2.2-i2v-fast | video | 260 credits — A very fast and cheap PrunaAI-optimized version of Alibaba's Wan 2.
- **[Wan 2.2 T2V Fast](https://console.core.today/llms/models/wan-video/wan-2.2-t2v-fast.txt)**: wan-video/wan-2.2-t2v-fast | video | 240 credits — A very fast and cheap PrunaAI-optimized version of Alibaba's Wan 2.

## Audio & TTS

- **[Bark](https://console.core.today/llms/models/suno-ai/bark.txt)**: suno-ai/bark | audio | 190 credits — Suno's text-prompted generative audio model.
- **[Chatterbox](https://console.core.today/llms/models/resemble-ai/chatterbox.txt)**: resemble-ai/chatterbox | audio | 59 credits — Resemble AI's production-grade open-source TTS with unique emotion exaggeration control and instant voice cloning from a short reference audio.
- **[Chatterbox Multilingual](https://console.core.today/llms/models/resemble-ai/chatterbox-multilingual.txt)**: resemble-ai/chatterbox-multilingual | audio | 7 credits — Chatterbox open-source TTS in 23 languages with instant voice cloning and emotion exaggeration control.
- **[Chatterbox Turbo](https://console.core.today/llms/models/resemble-ai/chatterbox-turbo.txt)**: resemble-ai/chatterbox-turbo | audio | 59 credits — Resemble AI's fastest open-source TTS without sacrificing quality.
- **[Gemini 3.1 Flash TTS](https://console.core.today/llms/models/google/gemini-3.1-flash-tts.txt)**: google/gemini-3.1-flash-tts | audio | 299 credits — Google Gemini 3.
- **[Incredibly Fast Whisper](https://console.core.today/llms/models/vaibhavs10/incredibly-fast-whisper.txt)**: vaibhavs10/incredibly-fast-whisper | audio | 10 credits — Whisper large-v3 optimized for speed (38M+ runs) — transcribes roughly 150 minutes of audio in under 100 seconds using batched inference.
- **[Suno Music V5.5](https://console.core.today/llms/models/suno/music-v5-5.txt)**: suno/music-v5-5 | audio | 140 credits — Suno's latest model, with the highest audio quality and richest arrangements.
- **[Suno Music V5](https://console.core.today/llms/models/suno/music-v5.txt)**: suno/music-v5 | audio | 140 credits — Suno V5 music generation with improved audio quality and musicality.
- **[Suno Music V4.5](https://console.core.today/llms/models/suno/music-v4-5.txt)**: suno/music-v4-5 | audio | 140 credits — Suno V4.
- **[MiniMax Music 2.6](https://console.core.today/llms/models/minimax/music-2.6.txt)**: minimax/music-2.6 | audio | 350 credits — Generate full-length songs or instrumentals (up to ~6 minutes) from a text prompt.
- **[MiniMax Music 2.5](https://console.core.today/llms/models/minimax/music-2.5.txt)**: minimax/music-2.5 | audio | 350 credits — Generate full-length songs (up to ~5 minutes) with vocals, lyrics, and rich instrumentation.
- **[MiniMax Music Cover](https://console.core.today/llms/models/minimax/music-cover.txt)**: minimax/music-cover | audio | 700 credits — Reimagine any song in a different style.
- **[Qwen3 TTS](https://console.core.today/llms/models/qwen/qwen3-tts.txt)**: qwen/qwen3-tts | audio | 47 credits — Alibaba's unified TTS with three modes: preset speakers (custom_voice), instant voice cloning from reference audio (voice_clone), and creating a brand-new voice from a text description (voice_design).
- **[Inworld Realtime TTS 2](https://console.core.today/llms/models/inworld/realtime-tts-2.txt)**: inworld/realtime-tts-2 | audio | 59 credits — Inworld's most expressive TTS with natural-language steering — place bracketed instructions like [speak quickly] before the text they apply to.
- **[Suno SFX V5](https://console.core.today/llms/models/suno/sfx-v5.txt)**: suno/sfx-v5 | audio | 29 credits — Suno V5 sound effect generation.
- **[MiniMax Speech 2.8 HD](https://console.core.today/llms/models/minimax/speech-2.8-hd.txt)**: minimax/speech-2.8-hd | audio | 237 credits — Ranked #1 on Artificial Analysis Speech Arena and Hugging Face TTS Arena.
- **[MiniMax Speech 2.8 Turbo](https://console.core.today/llms/models/minimax/speech-2.8-turbo.txt)**: minimax/speech-2.8-turbo | audio | 142 credits — Low-latency MiniMax Speech 2.
- **[MiniMax Speech 2.6 HD](https://console.core.today/llms/models/minimax/speech-2.6-hd.txt)**: minimax/speech-2.6-hd | audio | 237 credits — Studio-quality multilingual text-to-speech with nuanced prosody, emotion control, and premium voices for professional applications.
- **[MiniMax Speech 2.6 Turbo](https://console.core.today/llms/models/minimax/speech-2.6-turbo.txt)**: minimax/speech-2.6-turbo | audio | 142 credits — Fast multilingual text-to-speech with emotional control, optimized for real-time applications with low latency.
- **[MiniMax Speech-02-Turbo](https://console.core.today/llms/models/minimax/speech-02-turbo.txt)**: minimax/speech-02-turbo | audio | 142 credits — Low-latency text-to-speech model with multilingual support, emotional voice control, and 300+ voice options.
- **[Sonilo v1.1 Text to Music](https://console.core.today/llms/models/sonilo/v1.1/text-to-music.txt)**: sonilo/v1.1/text-to-music | audio | 520 credits — Sonilo v1.
- **[Inworld TTS 1.5 Max](https://console.core.today/llms/models/inworld/tts-1.5-max.txt)**: inworld/tts-1.5-max | audio | 24 credits — Inworld's highest-quality realtime TTS with under 200ms latency.
- **[Clova Voice TTS Premium](https://console.core.today/llms/models/ncp-clova/tts-premium.txt)**: ncp-clova/tts-premium | audio | 153 credits — NAVER Clova Voice Premium TTS with 108 voices across 6 languages.
- **[ElevenLabs Turbo v2.5](https://console.core.today/llms/models/elevenlabs/turbo-v2.5.txt)**: elevenlabs/turbo-v2.5 | audio | 119 credits — High-quality, low-latency ElevenLabs text-to-speech in 32 languages.
- **[ElevenLabs v3](https://console.core.today/llms/models/elevenlabs/v3.txt)**: elevenlabs/v3 | audio | 237 credits — ElevenLabs' most expressive text-to-speech model.
- **[ElevenLabs v2 Multilingual](https://console.core.today/llms/models/elevenlabs/v2-multilingual.txt)**: elevenlabs/v2-multilingual | audio | 237 credits — ElevenLabs Multilingual v2 text-to-speech in over 30 languages.
- **[Sonilo v1.1 Video to Sound Effects](https://console.core.today/llms/models/sonilo/v1.1/video-to-sound-effects.txt)**: sonilo/v1.1/video-to-sound-effects | audio | 310 credits — Sonilo v1.
- **[Suno Vocal Separation](https://console.core.today/llms/models/suno/vocal-separation.txt)**: suno/vocal-separation | audio | 120 credits — Separate a Suno-generated track into clean vocal and instrumental stems.
- **[MiniMax Voice Cloning](https://console.core.today/llms/models/minimax/voice-cloning.txt)**: minimax/voice-cloning | audio | 6980 credits — Clone any voice from a 10-second to 5-minute audio sample.
- **[Whisper](https://console.core.today/llms/models/openai/whisper.txt)**: openai/whisper | audio | 10 credits — OpenAI's Whisper large-v3 speech recognition (144M+ runs) — the standard for transcription.
- **[Whisper Diarization](https://console.core.today/llms/models/thomasmol/whisper-diarization.txt)**: thomasmol/whisper-diarization | audio | 7 credits — Whisper transcription with speaker diarization (8M+ runs) — returns who said what, with per-segment speaker labels and timestamps.
- **[XTTS-v2](https://console.core.today/llms/models/lucataco/xtts-v2.txt)**: lucataco/xtts-v2 | audio | 26 credits — Coqui XTTS-v2 multilingual voice-cloning TTS.

## LLM (Chat Completions)

- **[Claude Haiku 4.5](https://console.core.today/llms/models/claude-haiku-4-5.txt)**: claude-haiku-4-5 | llm | 1 credits — Fast, cost-effective model for everyday tasks.
- **[Claude Opus 5](https://console.core.today/llms/models/claude-opus-5.txt)**: claude-opus-5 | llm | 5 credits — Anthropic's latest flagship Opus model, with a 1M-token context window by default and 128K max output tokens.
- **[Claude Opus 4.8](https://console.core.today/llms/models/claude-opus-4-8.txt)**: claude-opus-4-8 | llm | 5 credits — Anthropic's most capable Opus-tier model, with a 1M-token context window (200K on some surfaces), 128K max output tokens, and knowledge cutoff to January 2026.
- **[Claude Opus 4.7](https://console.core.today/llms/models/claude-opus-4-7.txt)**: claude-opus-4-7 | llm | 5 credits — Anthropic's latest flagship model with reliable knowledge cutoff to January 2026 and 128K max output tokens.
- **[Claude Opus 4.6](https://console.core.today/llms/models/claude-opus-4-6.txt)**: claude-opus-4-6 | llm | 5 credits — Anthropic's most capable model.
- **[Claude Opus 4.5](https://console.core.today/llms/models/claude-opus-4-5.txt)**: claude-opus-4-5 | llm | 5 credits — Anthropic's most powerful model for highly complex tasks.
- **[Claude Sonnet 5](https://console.core.today/llms/models/claude-sonnet-5.txt)**: claude-sonnet-5 | llm | 4 credits — Anthropic's newest Sonnet model, tuned for the best balance of speed, cost, and intelligence.
- **[Claude Sonnet 4.5](https://console.core.today/llms/models/claude-sonnet-4-5.txt)**: claude-sonnet-4-5 | llm | 4 credits — Anthropic's most intelligent and capable Sonnet model.
- **[Claude Sonnet 4](https://console.core.today/llms/models/claude-sonnet-4.txt)**: claude-sonnet-4 | llm | 3 credits — Balanced Sonnet 4 model offering strong reasoning and coding abilities at an efficient price point.
- **[Gemini 3.5 Flash](https://console.core.today/llms/models/gemini-3.5-flash.txt)**: gemini-3.5-flash | llm | 2 credits — Google's latest Flash-tier workhorse model.
- **[Gemini 3.1 Flash Image Preview](https://console.core.today/llms/models/gemini-3.1-flash-image-preview.txt)**: gemini-3.1-flash-image-preview | llm | 500 credits — Gemini 3.
- **[Gemini 3.1 Flash Lite](https://console.core.today/llms/models/gemini-3.1-flash-lite.txt)**: gemini-3.1-flash-lite | llm | 1 credits — Stable release of the ultra-lightweight Gemini 3.
- **[Gemini 3.1 Flash Lite Preview](https://console.core.today/llms/models/gemini-3.1-flash-lite-preview.txt)**: gemini-3.1-flash-lite-preview | llm | 100 credits — Ultra-lightweight variant of Gemini 3.
- **[Gemini 3.1 Flash Live Preview](https://console.core.today/llms/models/gemini-3.1-flash-live-preview.txt)**: gemini-3.1-flash-live-preview | llm | 300 credits — Gemini 3.
- **[Gemini 3.1 Pro Preview](https://console.core.today/llms/models/gemini-3.1-pro-preview.txt)**: gemini-3.1-pro-preview | llm | 500 credits — Google's latest and most capable Gemini model in preview.
- **[Gemini 3 Flash](https://console.core.today/llms/models/gemini-3-flash.txt)**: gemini-3-flash | llm | 500 credits — Google's most advanced reasoning model with state-of-the-art multimodal understanding, PhD-level reasoning, and leading coding performance.
- **[Gemini 3 Pro Image Preview](https://console.core.today/llms/models/gemini-3-pro-image-preview.txt)**: gemini-3-pro-image-preview | llm | 500 credits — Google's premium image generation model within the Gemini 3 Pro family.
- **[Gemini 3 Pro Preview](https://console.core.today/llms/models/gemini-3-pro-preview.txt)**: gemini-3-pro-preview | llm | 4 credits — Google's most powerful Gemini model in preview.
- **[Gemini 2.5 Flash](https://console.core.today/llms/models/gemini-2.5-flash.txt)**: gemini-2.5-flash | llm | 1 credits — Google's fast and efficient model with built-in thinking capabilities.
- **[Gemini 2.5 Pro](https://console.core.today/llms/models/gemini-2.5-pro.txt)**: gemini-2.5-pro | llm | 3 credits — Google's most capable model with state-of-the-art reasoning and 1M token context.
- **[Gemini 2.0 Flash](https://console.core.today/llms/models/gemini-2.0-flash.txt)**: gemini-2.0-flash | llm | 1 credits — Google's fastest and most capable model.
- **[Gemini 2.0 Flash Lite](https://console.core.today/llms/models/gemini-2.0-flash-lite.txt)**: gemini-2.0-flash-lite | llm | 0.5 credits — Ultra-lightweight version of Gemini 2.
- **[Gemini Embedding 001](https://console.core.today/llms/models/gemini-embedding-001.txt)**: gemini-embedding-001 | llm | 0.1 credits — Google's text embedding model for generating vector representations.
- **[GPT-5.6 Luna](https://console.core.today/llms/models/gpt-5.6-luna.txt)**: gpt-5.6-luna | llm | 1 credits — The fast, low-cost tier of OpenAI's GPT-5.
- **[GPT-5.6 Sol](https://console.core.today/llms/models/gpt-5.6-sol.txt)**: gpt-5.6-sol | llm | 5 credits — The flagship tier of OpenAI's GPT-5.
- **[GPT-5.6 Terra](https://console.core.today/llms/models/gpt-5.6-terra.txt)**: gpt-5.6-terra | llm | 3 credits — The balanced tier of OpenAI's GPT-5.
- **[GPT-5.5](https://console.core.today/llms/models/gpt-5.5.txt)**: gpt-5.5 | llm | 5 credits — OpenAI's newest flagship model with a 1.
- **[GPT-5.4](https://console.core.today/llms/models/gpt-5.4.txt)**: gpt-5.4 | llm | 5 credits — OpenAI's newest flagship model with 1M context window and 128K output tokens.
- **[GPT-5.4 Mini](https://console.core.today/llms/models/gpt-5.4-mini.txt)**: gpt-5.4-mini | llm | 2 credits — Fast and cost-efficient variant of GPT-5.
- **[GPT-5.4 Nano](https://console.core.today/llms/models/gpt-5.4-nano.txt)**: gpt-5.4-nano | llm | 1 credits — Ultra-lightweight and fastest GPT-5.
- **[GPT-5.2](https://console.core.today/llms/models/gpt-5.2.txt)**: gpt-5.2 | llm | 4 credits — OpenAI's latest and most advanced GPT model.
- **[GPT-5.1 (2025-11-13)](https://console.core.today/llms/models/gpt-5.1-2025-11-13.txt)**: gpt-5.1-2025-11-13 | llm | 3 credits — Dated snapshot of GPT-5.
- **[GPT-5](https://console.core.today/llms/models/gpt-5.txt)**: gpt-5 | llm | 3 credits — OpenAI's latest flagship model.
- **[GPT-5 Mini](https://console.core.today/llms/models/gpt-5-mini.txt)**: gpt-5-mini | llm | 1 credits — Fast and efficient variant of GPT-5.
- **[GPT-5 Nano](https://console.core.today/llms/models/gpt-5-nano.txt)**: gpt-5-nano | llm | 1 credits — Ultra-fast and lightweight variant of GPT-5.
- **[GPT-4.1](https://console.core.today/llms/models/gpt-4.1.txt)**: gpt-4.1 | llm | 3 credits — OpenAI's most capable model for coding and instruction following.
- **[GPT-4.1 Mini](https://console.core.today/llms/models/gpt-4.1-mini.txt)**: gpt-4.1-mini | llm | 1 credits — A significant leap in small model performance.
- **[GPT-4.1 Nano](https://console.core.today/llms/models/gpt-4.1-nano.txt)**: gpt-4.1-nano | llm | 1 credits — OpenAI's fastest and cheapest model.
- **[GPT-4o](https://console.core.today/llms/models/gpt-4o.txt)**: gpt-4o | llm | 3 credits — OpenAI's flagship multimodal model.
- **[GPT-4o Mini](https://console.core.today/llms/models/gpt-4o-mini.txt)**: gpt-4o-mini | llm | 1 credits — Cost-effective, fast model with strong performance.
- **[GPT Audio Mini](https://console.core.today/llms/models/gpt-audio-mini.txt)**: gpt-audio-mini | llm | 1 credits — Lightweight multimodal model with native audio input/output capabilities.
- **[MiniMax M2.7](https://console.core.today/llms/models/MiniMax-M2.7.txt)**: MiniMax-M2.7 | llm | 1 credits — MiniMax's flagship M2-series language model, served through an OpenAI-compatible API.
- **[OpenAI o4-mini](https://console.core.today/llms/models/o4-mini.txt)**: o4-mini | llm | 2 credits — Fast, cost-effective reasoning model optimized for coding and STEM tasks.
- **[OpenAI o3-mini](https://console.core.today/llms/models/o3-mini.txt)**: o3-mini | llm | 2 credits — Efficient reasoning model that delivers strong performance at lower cost.
- **[OpenAI o1](https://console.core.today/llms/models/o1.txt)**: o1 | llm | 15 credits — OpenAI's most advanced reasoning model.