Runway Gen-2 is a powerful text-to-video model that enables you to create amazing clips from text descriptions. Today, I'm going to show you how to use this model to create anime videos that you can use for a variety of content.
I watch YouTube videos a lot, and I've noticed that some have started to use Runway Gen-2 to generate content for their videos. I believe that many will soon follow suit. Instead of including random images that you find on the internet, you can leverage a tool like Runway Gen-2 to create beautiful and original video clips.
It's especially useful to use this model when you want to create videos that you otherwise could never find or produce. Let's say that you have a YouTube channel in which you talk about ancient civilizations. There is only a small amount of content that already exists and you can include in your videos. But by using this AI model, you can make cool clips that will make your content more interesting to viewers.
Although Runway Gen-2 can produce both realistic videos and animations, today I'm going to focus solely on the latter since that's the nature of this topic. You can also use these animated video clips for music videos.
How to Write Runway Gen-2 Prompts for Anime Videos
Alright, I have a small confession to make. I usually use Runway Gen-2 at the beginning of each month and create many videos in a short period of time. After a few days, all my credits are gone and I have to wait almost a whole month to get to experiment with this AI model again. But that's okay because it gives me enough time to work on new ideas, and I also have other tools like Midjourney and Stable Diffusion to keep me busy.
Runway Gen-2 enables users to create video clips from both text and images. When I started using this tool, I preferred the image-to-video model. But as time went on, I realized that I could make far better videos if I structured my prompts properly.
Given that this article is all about anime videos, the most important thing to remember is that you should clearly instruct the model to create a video in this style. You can do this by using a variety of different keywords like 'anime style', 'anime show', 'anime clip', and so on. You'll notice that I use at least two of these keywords in each of my prompts. I do this to ensure there is no confusion.
Now, Runway Gen-2 is far from perfect. Remember that this is an innovative technology that has yet to reach the level text-to-image models have achieved. At the time of writing this article, there are still many errors this model can make when generating videos. There is no doubt in my mind that the model will get better over time.
I had to mention this so that you know what to expect when you're using this model. You shouldn't set your expectations too high. Nevertheless, you'll see that you can get some very decent clips by using this model. Here are some of the tips that you should follow when writing prompts for anime videos:
- You'll often get the best results when the prompt focuses on a single character.
- Intense movement sometimes doesn't look too good in Runway Gen-2 clips. It's often a good idea to avoid having a character running in a video clip.
- Make sure you mention all the information that's relevant to the clips you're creating. That can be anything from the lighting and camera movement to the background and setting.
- Use keywords that will better explain the scene you're trying to create.
These are not rules that are necessary to follow. Instead, they should serve only as advice that you can try out when you're making videos. There is no right or wrong way to write your prompts. In the end, you decide what you'll create and how you will do it.
14 Runway Gen-2 Prompt Examples for Anime Videos
I have written prompts and generated 14 anime-style videos for you today. I came up with a variety of characters to show you how the Runway Gen-2 model works and what you can expect when you're using it. Let's check out the videos I created.
prompt #1: anime show, anime scene, anime style, close-up shot of a male hero charging up his inner strength to release a concentrated beam of energy, high intensity scene
If you've watched Dragon Ball Z, you most likely figured out where I got the inspiration for this clip as soon as you read the prompt. There are many different ways that you can change this scene. You could say that the concentrated beam of energy is released into the sky. You can pick any setting that you want.
You basically have the freedom to create any scene that you want. In my prompt, I primarily focused on the main character. Also, make sure that you mention the gender of the character in your prompt. Remember that whenever you don't mention a particular detail, it will be randomized by the AI model.
prompt #2: anime art style, anime show, battle between good and evil in a fictional dystopian realm, high intensity scene, action packed scene, magic, destruction, survival of the fittest
One thing you might notice when you're using Runway Gen-2 is that when a generated video includes multiple characters, their faces usually don't look too good. It would be asking too much at this time if the faces were perfect. This model will get more powerful over time, so for now you should appreciate the fact that achieving this level of quality is possible.
There is a lot going on in this clip. It features a dystopian setting and it looks like there is a war going on. The level of detail in the video is quite impressive.
prompt #3: anime art style, anime show, a female healer with an enchanting green aura around her, set in a high fantasy world, high key lighting
The hands don't look too good in this video, but that's completely okay. Even text-to-image models often make errors when they're generating hands. I really like the background in this image, as it adds to the healer theme. What's also a nice touch is the fact that the character has green hair.
prompt #4: anime scene, anime style, sea captain on an old wooden ship going on an incredible adventure, seeking riches, colorful, wholesome
The fire effects in this video are an interesting choice. I'm not exactly sure where it came from since I didn't mention it in my prompt. I originally wanted to select a different video, but this one kind of stuck with me. I ended up liking the visual effect. It kind of symbolizes that inner fire you have when you're on a quest to achieve something amazing.
prompt #5: anime style, anime show, a character traveling through the cosmos, universe in the background, wondrous, ethereal
There are a lot of different ways that you can instruct the AI model to create characters. I usually keep my prompts simple in these articles because I want to produce consistent results. But I experiment with a lot of different ideas.
You can always instruct the model to have the character either facing towards the camera or looking away from it. If you don't include this information in your prompt, the model will randomly pick an outcome. In this example, I found it appealing to have the character looking away from the camera because it put more focus on the cosmic background.
prompt #6: anime show, anime art style, a villain who controls fire to bring destruction to the world, high key lighting, scary, evil, horror aesthetic
Not all people are heroes. That's why I have to include some villains in my prompts as well. What I find really interesting about this clip is how the facial expression of the character quickly changes. It goes from a slight smile to a devilish grin really quickly. The red eyes are also a nice touch, and the flames definitely add to the overall aesthetic.
prompt #7: anime style, anime, alchemist brewing a potion in a secret lab, medieval setting, fantasy world, magic
The objects featured in this video are quite interesting. It's like they're a blend of animation and realism. The facial expressions of the main character may seem a bit off at times, but I still appreciate the fact that there is movement in this clip. The fire animation is also very interesting.
prompt #8: anime, anime show, shaman in an ethereal jungle with glowing bioluminescent plants in the background, performing a sacred ceremony
Sometimes, I write something in my prompt on the off chance that the model actually generates what I described. I usually keep my expectations low, so I was pleasantly surprised when I saw these gorgeous bioluminescent plants. I also like the blue energy that's surrounding the shaman character.
prompt #9: anime style, anime clip, character that can control elements like fire, water, and air to help people, high fantasy setting
The background here is absolutely insane. It's like water and fire clashing. The character basically stands still but that's okay. This type of scene reminds me of an intro clip you'd see in an animated TV show where they introduce the characters.
prompt #10: anime style, anime, intergalactic explorer encountering a black hole for the first time, scary revelation, suspense
I'm a huge fan of Cristopher Nolan and the Interstellar movie. It's probably in my top 5 movies of all time, and it inspired me to write this prompt. I prayed that I'd get a good result from the prompt, and the generated video was way beyond any expectation that I had. The black hole imagery is mindblowing.
This is another instance in which the background is more important than the main character. It's also a good thing that the character is not looking at the camera so that the viewer can focus on the black hole. This is one of my favorite videos that I've created in Runway Gen-2. It's a good idea to add some music to clips like this to make the scene more dramatic.
prompt #11: anime art style, anime show, a skilled thief running through a cyberpunk dystopian megacity, close-up shot, intense, low key lighting
I previously mentioned in this article that it's hard to achieve intense movement in Runway Gen-2 that actually looks good. The movement usually doesn't look very natural. But let's take a minute to appreciate how the setting was generated. Look at all the details featured in this video. It's absolutely stunning.
prompt #12: anime clip, anime, a huge man guarding entrance to an ancient temple, desert setting
When I write prompts, I usually write a much longer description and then trim any details I don't consider important. In my initial description, I wrote that there is a source of energy in the middle of the temple. I ultimately decided to keep it simple, and that helped produce this great video. One of the best pieces of advice I can give is to keep your prompts as concise as possible. That's how you get consistent results.
prompt #13: anime style, anime, character with crystalline wings that protects people from evil, colorful, high fantasy setting
The wings here are spectacular. There is an insane level of detail featured in them. The armor the character is wearing is also quite impressive. I love it when an AI model includes details that I didn't mention in my prompt but add so much to the scene.
prompt #14: anime, anime style, vengeful ghost bound to an old haunted mansion looking for victims
This scene looks quite dark and grim. The ghost character is also quite scary. The candles on the left are also a nice touch. I would love to see the haunted mansion from a few different angles because it looks incredible.
Final Thoughts
While some of the videos above definitely stand out from the rest, I think that all of them are quite good. It's amazing that you can create these short clips from relatively simple descriptions.
Although it would arguably be very difficult to create a coherent video spanning at least several videos by piecing clips like these together, you can use these animations for a variety of online content. They can be used in anything from music videos to educational content.
One thing to remember is that you should keep your prompts concise. There is a 320-character cap in Runway Gen-2 when you're writing prompts for videos. I usually keep my prompts at a 180-character length or less.
