AI models

Daily check-in: 15 free credits

Model library

AI Video and Image Models on AgentReel

Pick a model to see its settings, its full credit price table and how it compares, then open the generator with it already selected.

Last updated

AI video models

AI image models

How to choose an AI video model

Start from the format you need

Every video model with a guide here renders 16:9 landscape and 9:16 vertical, but the other frames differ. If you need a square post, a 4:3 slide or a 21:9 cinematic strip, check the aspect ratio row of the settings table on the model page before you write a prompt. Seedance 2.5, Seedance 2.0 and MiniMax H3 cover six ratios including 21:9, Wan 3.0 and Grok Imagine Video 1.5 cover five (Grok adds 3:2 and 2:3), Kling 3.0 adds 1:1, and Veo 3.1 and Gemini Omni 1.1 Flash stay with landscape and vertical.

Decide how much control you want

Every video model with a guide accepts a text prompt on its own. Most also take up to two images used as the first and the last frame; Grok Imagine Video 1.5 takes one start image. A first frame fixes the subject and the look; a last frame fixes where the shot lands. Use frames for products, characters that must look the same across several clips and logo reveals, and plain text for B-roll and mood shots where a surprise is welcome. You can make the frames themselves with the AI image generator.

Sound, length and resolution

Veo 3.1, both Seedance models and Wan 3.0 generate sound with the picture, and Kling 3.0 lets you switch sound on or off. Veo 3.1 and Gemini Omni 1.1 Flash render 4K, MiniMax H3 reaches 2K, Grok Imagine Video 1.5 stops at 720p, and the others top out at 1080p. Clip length is fixed on Veo 3.1 and selectable on the rest, up to 15 seconds.

Match the model to the budget

Prices are credits per clip, and the range on each card above runs from the cheapest to the most expensive setting. Resolution moves the price most, and on every model with a length setting a longer clip costs proportionally more. Veo 3.1 is priced per clip by variant and resolution instead. A habit that saves credits: draft at the lowest resolution on a cheaper model, then render the keeper on the model and resolution you want to publish, all in the same AI video generator.

Credits, refunds and free use

Video models are paid from the credits of a plan or credit pack. The free credits a new account gets cover Z-Image, video templates and Code to Video, not the video models on this page. The Generate button shows the exact price before anything is charged, and a task that fails is refunded automatically. The credits and pricing guide explains how plans and packs differ.

Every model in the AI video generator

The generator offers 20 video models. Models with a guide link to it; every one of them can be picked in the generator.

ModelDeveloperCredits
Veo 3.1ProGoogle8–93
Seedance 2.5NewByteDance35–593
Kling 3.0HotKuaishou11–102
Kling 3.0 OmniKuaishou18–87
Gemini Omni VideoGoogle16–53
Gemini Omni 1.1 FlashNewGoogle16–53
Seedance 2.0ByteDance24–383
Wan 3.0NewAlibaba10–120
RunwayRunway3–8
MiniMax H3MiniMax12–49
Hailuo 2.3MiniMax8–23
Wan 3.0 PrimeAlibaba16–189
Kling 3.0 TurboFastKuaishou23–85
Seedance 2.0 FastFastByteDance15–93
PixVerse V6PixVerse7–69
HappyHorse 1.1HappyHorse29–109
Vidu Q4Vidu9–77
Wan 2.7Alibaba20–90
Grok Imagine Video 1.5xAI4–17
Grok Imagine VideoFastxAI4–30

Every model in the AI image generator

The image generator offers 20 image models. Models with a guide link to it; every one of them can be picked in the generator.

ModelDeveloperCredits
Nano Banana ProProGoogle5–6
Nano Banana 2.1NewGoogle1–3
Nano Banana 2Google2–5
Nano BananaFastGoogle1
GPT Image 2.5 SunburstNewOpenAI2–4
GPT Image 2.5 FlareNewOpenAI2–4
GPT Image 2OpenAI2–4
GPT Image 1.5OpenAI1–6
Seedream 5.0 ProHotByteDance2–4
Seedream 5.0 LiteByteDance2
Seedream 5.0 FlashFastByteDance1
Seedream 4.5ByteDance2
Flux 2 ProBlack Forest Labs2
Z-ImageFastAlibaba3
Flux KontextBlack Forest Labs2–3
Ideogram V3Ideogram1–3
Qwen Image 3.0 ProAlibaba2–3
Grok Imagine Image 2.0xAI1
Wan 2.7 Image ProAlibaba3
Imagen 4Google1–3