Models
Every top AI model, one workbench
Switch between the newest and most popular AI video and image models without switching tools. Compare strengths, then generate with any of them inside LazyKiwi.
OpenAI Sora 2 · Alternatives
Sora 2 alternative
OpenAI is retiring Sora 2, and LazyKiwi never ran it. Here is what to use instead, by job.
Explore modelMiniMax · Video Model
Hailuo 2.3 Fast
MiniMax Hailuo 2.3 Fast in LazyKiwi: image-to-video only, 6 to 60 seconds, 768p or 1080p, framed by your upload.
Explore modelMiniMax · Video Model
MiniMax H3
MiniMax's H3 video model — 768p or 2K output and the longest prompt field in LazyKiwi.
Explore modelKuaishou · Video Model
Kling v3 Omni
Kuaishou's Kling 3 generation: start-and-end frames, 3 to 60 second clips, 1080p output.
Explore modelAlibaba Wan · Video Model
Wan 2.7
Alibaba's Tongyi Wanxiang video model in LazyKiwi: one still or a keyframe pair in, 720p or 1080p out.
Explore modelAlibaba Wan · Video Model
Wan 2.2
Alibaba’s Wan2.2 MoE I2V models — short, controlled motion from a still in LazyKiwi.
Explore model
Seed · Video Model
Seedance 2.0
Seed's multimodal AI model for controllable video generation and editing from text and images.
Explore modelByteDance Seed · Video Model
Seedance 2.5
ByteDance Seed's newest video model — up to 30 seconds a clip, 21:9 through 9:16.
Explore modelLightricks · Video Model
LTX-2.3
Lightricks’ DiT foundation model for synchronized video — available in LazyKiwi with promo pricing.
Explore modelTencent · Video Model
HunyuanVideo 1.5
Tencent’s 8.3B open video model for coherent motion — T2V and I2V in LazyKiwi.
Explore modelAlibaba · Video Model
HappyHorse 1.0
Alibaba's HappyHorse 1.0 in LazyKiwi: text-to-video and image-to-video, five aspect ratios, 3 to 60 seconds, 1080p by default.
Explore modelxAI · Video Model
Grok Imagine Video
The Grok video generator with the widest ratio set in LazyKiwi, 21:9 to 4:5.
Explore modelGoogle DeepMind · Video Model
Veo 3.1 Lite
Google DeepMind's Veo 3 generation, Lite tier: 16:9 or 9:16, 720p or 1080p, 4 to 60 seconds.
Explore model
OpenAI · Image Model
GPT Image 2.5
OpenAI's newest image model, in both variants, in one browser workbench.
Explore model
Seed · Image Model
Seedream 5.0
Seed's flagship text-to-image model for professional visual content creation
Explore model
Google · Image Model
Nano Banana
Google's Gemini image model, in your browser
Explore model
OpenAI · Image Model
GPT Image 2
OpenAI's advanced image generation model for detailed, accurate visual content creation.
Explore model
Alibaba · Image Model
WAN 2.7 Image Pro
Alibaba's enterprise-grade image generation model for creative and commercial workflows
Explore model
Tongyi-MAI · Image Model
Z-Image
Alibaba Tongyi-MAI’s 6B S3-DiT image model — fast text-to-image in LazyKiwi.
Explore model
Qwen · Image Model
Qwen Image
Qwen’s Apache 2.0 text-to-image model — generate or edit a single image in LazyKiwi.
Explore model
Image
Nano Banana 2.1
Google's Gemini image model from October 2026: sharper text, 4K output, up to 14 reference images.
Explore model
Image
Qwen Image 2.1 Turbo
Qwen's 8-step distilled version of Qwen-Image-2.1 for fast local generation.
Explore model
Text
Gemini 4 Argon
Google's frontier text and reasoning model, announced September 2026 with limited access.
Explore model
Text
GPT-6 Astra
OpenAI's most capable model for coding, computer use and professional work, released September 2026.
Explore model
Video
Ray 3
Luma's reasoning video model with 16-bit HDR output; Ray3.2 adds up to 16 keyframes and an API.
Explore model
Video
Kling 4.0
Kuaishou's September 2026 model: native 30-second clips, 4K 10-bit HDR and multiple keyframes.
Explore model
Image
FLUX.2
Black Forest Labs' image family from November 2025: pro, flex, 32B open-weight dev, klein and max.
Explore model
Image
Imagen
Google DeepMind's text-to-image line; Imagen 4 (May 2025) was retired in 2026 in favour of Nano Banana.
Explore model
Image
Stable Diffusion
Stability AI's open-weight image family: 3.5 Large (8.1B), Large Turbo and Medium (2.5B), free under $1M revenue.
Explore model
Video
Kling 1.6
Kuaishou's advanced Kling ai video generator model with extended duration and motion control
Explore model
Video
Grok 2
xAI's advanced video generation model for creators and developers
Explore modelPick a model and start creating.
Every model above runs in the same workbench, with no extra tools and no waitlists.
Open the workbench