Quick CreateGenerate instantly with multiple top models in one place.
Model APIsHundreds of models with fast, reliable API access.
Online ComfyUIUse ComfyUI online with ready-made models, workflows, and apps.
Google Nano Banana 2.1 generates and edits images in one place, covering text-to-image and multi-image reference editing with fusion, up to 4K output, 15 aspect-ratio choices, 1-4 images per request, and optional web search grounding, billed per image by resolution tier.
Google Nano Banana 2.1 is the channel edition of Google's image model, generating and editing images at Flash-level speed with improved quality, in-image text rendering, and multi-turn consistency, supporting up to 14 reference images, 1K/2K/4K output, and 14 aspect ratios. Priced significantly lower than the official edition, billed per call by resolution tier.
GPT Image 2.5 Sunburst is OpenAI's precision-focused image tier supporting text-to-image generation and multi-reference editing, with five quality tiers, transparent background, and accurate in-image text. Billed per output image by quality and image size, plus per additional input image.
GPT Image 2.5 Sunburst (channel edition) is OpenAI's image generation tier for text-to-image and multi-reference editing via a low-price channel, with flat per-call pricing and flexible resolutions and aspect ratios. Priced lower than the official edition.
GPT Image 2.5 Flare is OpenAI's fast, balanced image tier supporting text-to-image generation and multi-reference editing, with five quality tiers, transparent background, and accurate in-image text. Billed per output image by quality and image size, plus per additional input image.
GPT Image 2.5 Flare (channel edition) is OpenAI's fast, balanced image tier for text-to-image and multi-reference editing via a low-price channel, with flat per-call pricing and flexible resolutions and aspect ratios. Priced lower than the official edition.
Seedream 5.0 Pro is ByteDance's flagship image generation and editing model, supporting text-to-image, single-reference image editing, and layer decomposition with output up to 2K resolution. Output is billed per image (or per generated layer) by pixel tier, suited to posters, brand visuals, and professional design workflows.
Qwen Image 3.0 is Alibaba's image generation and editing model with native multilingual prompting and best-in-class text rendering, capable of multi-line layouts and paragraph-level typography in images. It supports text-to-image and editing with 1-3 reference images, billed per output image.
Qwen Image 3.0 Pro is the premium edition of Alibaba's image model, tuned for higher-quality editing and transformation with up to 4.5k-token prompts, 12 languages, and 20+ fonts for native text rendering in complex multi-line layouts. It supports text-to-image and editing with 1-3 reference images; output is billed per image by size tier, plus $0.003 per input image in edit mode.
Grok Imagine Image 2.0 is xAI's image generation and editing model, creating detailed images from text prompts or reference images, with quality tiers (low/medium) at 1K/2K resolution. Billed per image.
ByteDance Seedream 5.0 is a new-generation text-to-image model focused on knowledge reasoning and intelligent editing, and the first Seedream release with real-time web retrieval. It offers deeper world-knowledge understanding, improved semantic comprehension and spatial-logical reasoning for complex prompts, with native 4K output for infographics, knowledge-reasoning illustrations, and e-commerce design. Edit mode supports multi-turn editing via natural language without regeneration.
Google Nano Banana 2 is a new-generation image generation model. Inheriting the advanced intelligence of the Pro edition while delivering faster generation, it supports output from 512px up to 4K. It achieves significant gains in text rendering, world-knowledge integration, character consistency, and instruction adherence, and can directly produce finished-grade layouts for infographics, multilingual menus, and dynamic illustrations.
Google Nano Banana 2 is a high-performance multimodal image-generation base model. It supports direct text-to-image generation of high-fidelity images, or precise controllable generation with reference images, with outstanding visual expression and detail fidelity. The channel edition is more economical than the official, suitable for social-media assets, marketing collateral, and e-commerce product images. Its edit mode performs stable editing at high speed, enhanced by real-time search.
Google Nano Banana Pro is a professional image generation model supporting up to 4K output, leveraging advanced physical light-shadow rendering and text understanding to generate industrial-grade visuals from precise instructions. The channel edition is priced lower than the official, suitable for daily creative scenarios. Its edit mode supports up to 14 reference images with control over lens angle, focal length, color grading, and scene lighting.
Google Nano Banana Pro is a professional image generation model supporting text-to-image generation and multi-image reference fusion, with ultra-clear rendering and up to 4K output. It features advanced semantic understanding and structured reasoning, and excels at complex instruction understanding, detail presentation, and precise multi-language text rendering. Suitable for brand advertising, product posters, and other visual content with strict text-presentation requirements.
Alibaba Wan 2.7 Image is the standard-tier text-to-image model with an upgraded core rendering architecture and precise text-semantic understanding. It offers coordinated composition, soft, layered color rendering, and compatibility with realistic, anime, and minimalist styles, with custom size control, built-in thinking mode, and broad aspect-ratio support. Edit mode preserves original composition and subject identity, supporting style transfer, quality enhancement, and fine-tuning.
OpenAI GPT Image 2 turns natural-language prompts into high-quality images. It supports text-to-image generation and multi-image reference editing, excelling at complex instruction understanding, detailed rendering, and precise multi-language text layout. The official edition adds a reasoning-verification mode and can output up to 8 stylistically consistent images per prompt.
OpenAI GPT Image 2 is an image-generation interface built on a frontier large-model architecture. It supports deep text-semantic understanding, accurately captures complex prompt details, and transforms them into high-quality, well-composed visuals. The channel edition is priced lower than the official, suitable for daily visual and marketing collateral scenarios. Its edit mode maintains character appearance, lighting, and subject identity without manual masking for pixel-level edits.
Alibaba Wan 2.7 Pro Image is the professional tier of the Wan 2.7 text-to-image model, supporting up to 4K (4096x4096) output with built-in thinking mode and custom size control. It delivers higher-fidelity compositions with precise proportions, rich scene detail, and high-end light-shadow, compatible with realistic, Chinese-style, and anime styles for commercial visual design and concept art. Edit mode preserves composition via intelligent style remodeling.
ByteDance Seedance 2.5 US is the overseas edition of ByteDance's frontier cinematic video model, generating video with native audio from text, reference materials, or first- and last-frame images. Optimized for global access with lower latency outside mainland China, it is billed per second by resolution tier.
Seedance 2.5 is ByteDance's latest-generation video model supporting up to 4K output with native audio, cinematic camera control, and multi-character consistency. It covers generation from text, first-last frames, or multimodal references as well as prompt-driven video editing and extension, billed per second of output video by resolution tier.
Google Gemini Omni Flash 1.1 is a lightweight multimodal video model that turns a text prompt, reference materials, or first- and last-frame images into 3-10 second videos with synchronized audio, and edits existing footage via natural language. It supports 360p to 4K in landscape or portrait, billed per second by resolution tier.
MiniMax H3 (channel edition) is the economy edition of the MiniMax H3 video model, generating video from text prompts, first and last frames, or reference images, videos, and audios at 2K or 768P resolution. Billed per second of output video by resolution tier, suited for high-frequency, cost-sensitive production scenarios.
MiniMax H3 is MiniMax's multimodal video model that generates video from text prompts, first and last frames, or reference images, videos, and audios, at up to 4K resolution with strong prompt understanding and cinematic quality, billed per second of output video by resolution tier.
Wan 3.0 is Alibaba's video generation model producing high-quality video from text prompts, reference images or videos, or first and last frames, at 480P/720P/1080P with flexible duration and aspect ratio options. Billed per second of output video by resolution tier; uploaded reference videos are also billed per input second.
ByteDance Seedance 2.0 is a new-generation native multimodal video-generation model supporting joint input of text, images, video clips, and audio. With powerful audio-visual sync and motion perception, it locks onto character visuals, camera movement, and audio rhythm via @-tags, outputting film-grade high-fidelity video. The channel edition supports text-to-video, first-last-frame, and reference-to-video generation with up to 9 images / 3 videos / 3 audios as reference.
Happy Horse 1.1 is Alibaba's video generation model (v1.1), generating cinematic 480p/720p/1080p video from text, reference images, or a first-frame image, with smooth camera movement and expressive motion. Billed per second of output video by resolution tier.
ByteDance Seedance 2.0 is ByteDance's flagship text-to-video model, deeply tuned for full-scenario general generation quality. It generates high-fluency, physically consistent video from text or images, with native binaural audio, excellent multi-shot continuity and detail rendering. It autonomously plans storyboards and maintains strong subject consistency across multi-character, multi-plot narratives, making it a key tool for film studios and ad teams delivering high-quality finished clips.
ByteDance Seedance 2.0 Fast is the official accelerated text-to-video variant of the Seedance 2.0 multimodal audio-video joint generation model on Seed's unified architecture. It supports mixed input of text, image, audio, and video, delivers strong physical realism and fine instruction control, and generates high-quality multi-shot audio-video content with native binaural audio at low latency for commercial teams needing stable service and standard quality.
ByteDance Seedance 2.0 Fast is a film-grade video model optimized for faster, lower-cost generation on Seed's unified multimodal architecture. With a powerful physics engine and aesthetic understanding, it recreates shot design, motion changes, and camera rhythm from text, images, or audio, suiting advertising, creative shorts, and film pre-visualization. The channel edition compresses multimodal inference overhead for batch production while keeping brand-visual alignment and audio-video sync.
Built on the Omni One architecture, Kuaishou Kling O3 Pro supports text-to-video generation and precise modification of existing videos via natural-language instructions — covering scene atmosphere, element substitution, and lighting style — while preserving original motion trajectories and character continuity without manual masking. The channel edition is priced lower than the official version, suiting creative short videos, ad-script visualization, and film-concept dynamic previews.
Alibaba Happy Horse 1.0 generates cinematic 720p/1080p videos from text or a reference image, with smooth camera movement, expressive motion, and strong prompt fidelity. The official edition is tuned for ad spots, short-drama segments, and scenarios demanding higher visual quality, with up to 15-second multi-shot narration and native audio-visual sync.
Alibaba Wan 2.7 Video is an advanced video generation model supporting five modes: text-to-video, image-to-video, reference-to-video, video editing, and video extension. It relies on high-precision semantic understanding to produce high-definition dynamic footage with natural character motion, smooth camera movement, and rich detail, handling complex scenes and multi-character narratives for commercial, film-short, and artistic-video workflows.
Kwaivgi Kling Create Voice creates a custom voice ID from a 5 to 30 second clean single-voice audio clip, for use with Kling voice control in video generation. Supports mp3/wav audio or mp4/mov video input. Billed per voice creation request.
Kling 4.0 Flash Video Edit is Kwaivgi's video-editing model: give it an input video (@Video1) and prompt instructions describing the edits shot by shot (up to 8 shots with start and end timestamps), and it regenerates the footage to follow the new script while keeping the narrative structure of the original. 720p output; output duration defaults to the source clip's length; audio is generated natively.
Kling 4.0 Flash Text to Video is Kwaivgi's fast text-to-video flagship: describe any idea and it produces coherent 720p footage with sharp prompt following and fluent physics. Duration 3-20 seconds; aspect ratio auto/16:9/9:16/1:1/21:9 (auto adapts to the prompt); audio is generated natively.
Kling 4.0 Flash Reference to Video is Kwaivgi's multi-reference video model: generate a coherent 720p video from a prompt plus multi-reference assets, locking subject identity across shots. Up to 10 reference images, up to 5 reference videos (each 3-20s, combined with element videos within a 20s budget), and up to 7 elements (a frontal image, 1-3 multi-angle references, or a 3-8s element video); voices not accepted. Aspect ratio auto/16:9/9:16/1:1/21:9; audio generated natively.
Kling 4.0 Flash Image to Video is Kwaivgi's image-to-video model that takes a single image as the first frame and animates it: the prompt describes the shot (Flash does not take an end frame). Duration 3-20 seconds, 720p only, native audio generation.
GPT Image 2.6 Sunburst Text to Image is OpenAI's precision-focused image model, built for premium visual work with extra fidelity on intricate details, in exchange for longer generation times. Quality has four tiers (auto/low/medium/high) with a separate reasoning_effort control; output supports JPEG/PNG/WebP/GIF and SVG; up to 10 images per call.
GPT Image 2.6 Sunburst Image to Image is OpenAI's editing model built for the tightest control: edits scoped precisely to the instruction, with subject and composition preserved across many rounds of revision. Up to 16 reference images; mask or markup (annotated overlay) local editing; a frame pipeline renders stills to a single GIF loop (SVG unavailable here). Quality has four tiers plus a separate reasoning_effort control.
GPT Image 2.6 Flare Text to Image is OpenAI's image model for most applications: fast, high-quality generation with natural lighting, rich textures, and support for complex layouts including transparent backgrounds. Quality has four tiers (auto/low/medium/high) with a separate reasoning_effort control; output supports JPEG/PNG/WebP/GIF and SVG; up to 10 images per call.
GPT Image 2.6 Flare Image to Image is OpenAI's fast, high-quality editing model: it changes only what is asked, keeping subject, composition, and background intact, with reference subjects staying recognizable across successive edits. Up to 16 reference images; mask or markup (annotated overlay) local editing; a frame pipeline renders stills to a single GIF loop (SVG unavailable here). Quality has four tiers plus a separate reasoning_effort control.
Google Gemini Omni Flash 1.2 Video Upscale upscales a video of up to 15 seconds to a selected resolution of up to 4K, with an optional text instruction to guide the result. The target resolution must be higher than the source resolution.
Google Gemini Omni Flash 1.2 Video Extend continues an existing video from a text prompt, conditioning on its last 10 seconds and appending a new 3 to 10 second segment. Output resolution is selectable from 360p, 720p, 1080p, or 4K.
Google Gemini Omni Flash 1.2 Video Edit edits an existing video of up to 15 seconds from a natural-language instruction, applying the requested change while keeping the rest of the scene unchanged. Output resolution is selectable from 360p, 720p, 1080p, or 4K.
Google Gemini Omni Flash 1.2 Text to Video is Google's multimodal video model that generates video with synchronized native audio from a text prompt. Output is selectable at 360p, 720p, 1080p, or 4K in 16:9 or 9:16, with durations from 3 to 15 seconds.
Google Gemini Omni Flash 1.2 Reference to Video generates video from a text prompt combined with up to ten reference images and up to three reference videos, each addressable in the prompt as visual references for subjects and settings. Output is selectable at 360p, 720p, 1080p, or 4K, with durations from 3 to 15 seconds.
Google Gemini Omni Flash 1.2 FLF to Video animates a start image into video with synchronized native audio, with an optional end image for first-to-last-frame interpolation of up to 10 seconds. Output is selectable at 360p, 720p, 1080p, or 4K, with durations from 3 to 15 seconds.
BizyAirPlus
Run ComfyUI on Cloud GPUS
No local GPU needed. Install the plugin and run any workflow in the cloud.
Get the PluginAfter installation, start ComfyUI and choose BizyAirPlus mode from the run button to begin.
