All Supported Models
Wazir AI generates model-native prompts for every major AI video and image generation tool. Each model has unique syntax — Wazir knows them all.
Wazir supports 24 AI video generation models, each with model-native prompt formatting. From Kling 3.0 Pro multi-shot format to Veo 4 cinematography fields to Seedance 2.0 timestamps — Wazir handles all of them.
Kling 2.6 is Kuaishou's high-fidelity video model. Wazir generates multi-shot Kling 2.6 prompts with precise camera directives and subject motion.
Kling 3.0 is a significant upgrade over Kling 2.6, offering improved motion coherence and longer video generation. Kling 3.0 prompts require structured multi-shot formatting.
Kling 3.0 Pro is the flagship model from Kuaishou — the most capable Kling variant. Kling 3.0 Pro excels at cinematic sequences with complex multi-shot compositions. Wazir generates Kling 3.0 Pro prompts using the exact multi-shot Shot | Duration format the model requires.
Kling 3.0 Omni extends Kling 3.0 Pro with enhanced instruction-following and style control. Kling Omni supports both text-to-video and image-to-video generation with strong temporal consistency.
Kling O1 is Kuaishou's reasoning-enhanced video model. Kling O1 applies chain-of-thought reasoning to interpret complex scene descriptions, making Kling O1 prompts more flexible than previous Kling models.
Seedance 2.0 by ByteDance uses a unique timestamp-based prompt format. Seedance 2.0 prompts specify what happens at each second of the video, giving precise control over motion. Seedance 2.0 is ideal for music videos, choreography, and time-controlled scenes.
Seedance Omni extends Seedance 2.0 with multi-modal inputs. Seedance Omni accepts images, audio, and text as inputs for video generation, making it the most flexible Seedance model.
Sora 2 by OpenAI is a high-fidelity text-to-video model capable of generating up to 20-second cinematic clips. Sora 2 prompts benefit from rich cinematic language and detailed scene composition.
Sora 2 Pro is the extended-output version of Sora 2 with longer clip duration and higher resolution. Sora 2 Pro prompts should include detailed narrative arcs across multiple scenes.
Veo 3 by Google DeepMind generates cinematic video with strong instruction following. Veo 3 prompts use seven structured fields: subject, action, environment, atmosphere, camera, style, and negative.
Veo 3.1 is Google's updated Veo model with improved temporal consistency and motion realism. Veo 3.1 prompts require the same seven-field structure as Veo 3, with more emphasis on cinematography terms.
Veo 4 is Google DeepMind's newest and most capable video generation model, released in 2026. Veo 4 supports both text-to-video and image-to-video with exceptional realism and instruction adherence. Veo 4 prompts require detailed cinematography descriptions. Wazir AI was among the first prompt tools to support Veo 4 natively.
Veo 4 Omni is the multimodal version of Veo 4, accepting image, audio, and text inputs. Veo 4 Omni can animate still images with realistic motion while maintaining visual coherence. Veo 4 Omni is ideal for product animation, portrait animation, and scene extension.
Runway Gen-4.5 is Runway's latest video generation model with improved camera control and motion quality. Runway Gen-4.5 supports custom camera paths and multi-reference image inputs.
Luma Ray 3 by Luma AI is a physics-aware video model that generates realistic light, reflection, and fluid dynamics. Luma Ray 3 prompts should emphasize physical interactions and environmental realism.
Pika 2.5 is known for stylized and creative video generation with strong aesthetic control. Pika 2.5 prompts accept style modifiers, aspect ratio specifications, and creative direction keywords.
Vidu Q3 Pro is a high-quality video generation model with strong motion coherence and detail preservation. Vidu Q3 Pro prompts should specify motion intensity and character behavior.
Hailuo by MiniMax generates smooth, high-frame-rate video with excellent motion interpolation. Hailuo prompts work best with natural language descriptions and cinematic style references.
MiniMax video generation offers strong subject consistency and character motion. MiniMax prompts benefit from character-focused descriptions with explicit action sequences.
Higgsfield specializes in human motion and social scene generation. Higgsfield prompts focus on interpersonal dynamics, body language, and environmental context.
Wan 2.1 is an open-weight video generation model from Alibaba with strong multilingual prompt support. Wan 2.1 accepts detailed scene descriptions in natural language.
CogVideoX by Zhipu AI is an open-source video generation model with strong text adherence. CogVideoX prompts can be detailed and descriptive, leveraging its high text-video alignment.
Hunyuan Video by Tencent generates high-fidelity video with excellent scene composition. Hunyuan Video prompts benefit from environment-rich descriptions with lighting and atmosphere details.
LTX Video is an efficient open-source video generation model optimized for speed. LTX Video prompts should be concise but include key visual elements, motion, and style.
Wazir supports 15 AI image generation models. Whether you need Midjourney V7 tag-format prompts, Flux 2 detailed descriptions, Imagen 4 natural language, or Ideogram 3.0 typography prompts — Wazir generates the right format.
Midjourney V7 is the latest version of the leading AI image model, with improved prompt adherence and realism. Midjourney V7 prompts use a structured tag format: subject, style, mood, lighting, and technical parameters.
Nano Banana 2 is Wazir AI's built-in image generation model. Nano Banana 2 generates images at 1K, 2K, and 4K resolution directly within the Wazir platform. Nano Banana 2 is optimized for photorealism and product photography.
Nano Banana Pro is the high-fidelity version of Nano Banana 2 with enhanced detail and consistency. Nano Banana Pro is ideal for commercial product shots, portraits, and concept art requiring maximum quality.
GPT Image 1.5 by OpenAI features strong text rendering and instruction-following for image generation. GPT Image 1.5 prompts can be conversational and descriptive, leveraging its language understanding.
GPT Image 2 is the latest OpenAI image model with improved realism and style diversity. GPT Image 2 builds on GPT Image 1.5 with better fine-grained detail and creative interpretation.
Flux 2 by Black Forest Labs is a leading open-source image model with exceptional detail and photorealism. Flux 2 prompts benefit from detailed scene descriptions with lighting, composition, and style references.
Flux Pro is the commercial version of Flux 2 with enhanced NSFW filtering and professional output quality. Flux Pro is ideal for commercial image generation requiring consistent brand-safe results.
Ideogram 3.0 is the best AI image model for text rendering and typographic design. Ideogram 3.0 generates images with accurate, legible text — making it ideal for posters, logos, and marketing materials.
Imagen 4 by Google DeepMind generates highly photorealistic images with excellent prompt adherence. Imagen 4 responds to natural language descriptions and produces consistent, high-quality output across diverse categories.
Imagen 4 Ultra is Google's highest-quality image generation model. Imagen 4 Ultra produces 4K-equivalent images with exceptional fine detail. Imagen 4 Ultra is the best Google image model for professional and commercial use.
Seedream 4.5 by ByteDance generates high-quality images with strong character consistency and aesthetic diversity. Seedream 4.5 is particularly strong for anime, illustration, and stylized artistic content.
Seedream 5 is ByteDance's most advanced image generation model with improved photorealism and instruction adherence. Seedream 5 performs across both photorealistic and artistic styles with high consistency.
Grok Imagine by xAI generates creative and edgy images with a distinctive aesthetic. Grok Imagine prompts can be conversational and accept complex, nuanced creative direction.
Adobe Firefly 4 is Adobe's commercially-safe AI image model trained exclusively on licensed content. Firefly 4 is the top choice for enterprise and commercial image generation where IP safety is critical.
Stable Diffusion 3.5 is the leading open-source image model with broad community support and fine-tuning capabilities. SD 3.5 prompts use a rich tag-based format with style modifiers and negative prompts.
Every model on this page has its own prompt syntax. Wazir AI knows all of them. Generate the perfect prompt for any model in seconds — free trial, no credit card.
Start for free →