FLUX.2 [klein] 9B is a 4-step distilled image generation and editing model designed for sub-second inference without sacrificing visual quality. It unifies text-to-image and advanced editing workflows in a single model, making it suitable for interactive applications, real-time previews, and latency-critical production use.

ImagineArt 1.5 Pro is a high-resolution AI image generation model that creates native 4K visuals from text prompts and reference images. It focuses on enhanced realism, accurate text rendering, strong visual composition, and color placement consistency to support professional creative workflows such as poster design, product imagery, and branding assets.

FLUX.2 [klein] 4B Base is a compact undistilled image generation and editing model with an exceptional quality-to-size ratio. It is well suited for local deployment, efficient fine-tuning, and custom pipelines that require flexibility on limited hardware.

FLUX.2 [klein] 4B is a 4-step distilled image generation and editing model optimized for ultra-low latency inference. It delivers near real-time performance with strong visual quality, enabling interactive workflows and responsive production systems on more constrained hardware.

FLUX.2 [klein] 9B Base is the undistilled foundation model of the Klein family, offering full model capacity for image generation and editing. It is optimized for fine-tuning, customization, and post-training workflows where flexibility, control, and maximum training signal are required.

Qwen-Image-2512 is an improved version of the Qwen-Image image foundation model with enhanced prompt understanding, superior text rendering accuracy, and more realistic visual details. It generates high-fidelity images from text prompts across diverse subjects and styles.

Seedance 1.5 Pro is a next-generation AI video model from BytePlus that generates cinematic videos with native synchronized audio directly from text or image inputs. It offers precise audio-visual timing, strong motion coherence, expressive camera control, and advanced narrative prompt handling for short video creation.

Qwen-Image-Layered decomposes a static image into multiple RGBA layers, enabling independent editing of semantically distinct components without interfering with other parts of the image. This layered representation supports high-fidelity image editing tasks like resizing, repositioning, recoloring, and object manipulation with consistent detail and transparency handling.

Bria Video Eraser is a video editing model that removes objects from existing video using point-based selection, text instructions, or uploaded masks. It is designed to maintain temporal consistency across frames, prevent flickering and drift, and preserve the original audio track while modifying only the targeted visual regions.

Wan2.6 is a multimodal video model for text to video and image to video generation with support for multi-shot sequencing and native sound. It emphasizes temporal stability, consistent visual structure across shots, and reliable alignment between visuals and audio in short form video generation.

FLUX.2 [max] is a high-precision text to image and image editing model from Black Forest Labs that generates visuals grounded in real-time information via live web search. It delivers maximum prompt adherence with multi-reference editing and state-of-the-art consistency across identities, objects, and details.

GPT Image 1.5 is OpenAI’s newest flagship image model powering the latest ChatGPT Images. It delivers significantly faster image generation with stronger instruction following, more precise edits that preserve original details, more believable transformations, and improved rendering of dense or small text. It is suited for practical creative workflows, detailed design tasks, and production use cases.

react-1 is a video performance editing model designed for post-production direction without reshoots. It modifies acting and emotional delivery within existing footage while preserving identity and visual continuity, enabling directors to reshape performances using audio or written guidance.

Kling VIDEO 2.6 Pro is a full audio-visual AI video model that combines cinematic-quality video generation with native audio (dialogue, sound effects, ambience). It supports flexible workflows from text or image input, delivering synchronized video and sound in one pass with strong consistency and creative control. Via the API, Motion Control enables creators to guide character movement using a reference video for more realistic and physically grounded motion.

KlingAI Avatar 2.0 Pro builds on the Standard version with higher visual fidelity, smoother motion, and improved expressivity. It generates up to five-minute avatar videos from a single image and audio track, with enhanced detail and production-ready results for varied character types.

KlingAI Avatar 2.0 Standard generates talking avatar videos from a single portrait image and audio, preserving identity and producing natural lip-sync and expressive motion. It supports up to five minutes of video with multilingual control and gesture clarity for human or cartoon characters.

Seedream 4.5 is a ByteDance image model for precise 2K to 4K generation and editing. It improves multi image composition, preserves reference detail, and renders small text more reliably. It supports up to 14 reference images for stable characters and design heavy layouts.

Kling IMAGE O1 is a high control image generation model for stable characters and precise edits. It supports detailed composition control, strong style handling, and localized modifications without structural drift. Ideal for pipelines that need repeatable shots and complex visual continuity.

P-Image is a real-time text-to-image model from Pruna. It delivers sub-second image generation with strong text rendering and tight prompt adherence. It targets production workloads that need fast inference, predictable output control, and efficient scaling through simple API integration.

Kling VIDEO O1 is a unified multimodal video foundation model for controllable generation and instruction based editing. It supports text prompts, visual references, and video input so developers can build high control pipelines for pacing, transitions, object changes, and style revisions.

Kling VIDEO O1 Pro is a unified multimodal video foundation model for controllable generation and instruction based editing. It supports text prompts, visual references, and video input so developers can build high control pipelines for pacing, transitions, object changes, and style revisions.

PixVerse v5.5 is a director focused video model for story driven clips. It supports multi image fusion for character continuity, multi shot sequences, and native audio. It delivers smooth motion, refined cinematic control, and precise text guided video generation for complex scenes.

LTX-2 Retake regenerates targeted segments inside an existing clip. Define start time and duration. Replace video, audio or both with new prompts. Preserve surrounding motion and continuity. Ideal for selective revisions, dialog tweaks, and shot refinements without full regen.