Pictures, clips and songs
Ask for a picture, a video clip, a song, a voice or a sound effect in the same box: how to ask, what each costs, how long it takes and where it goes.
On this page
Ask for it
Describe what you see in your head: the subject, the setting, the style, the light, the mood. A short line is fine: askr can fill in the rest, and when it does, More, Details, Prompt used shows the exact words the model was given.
Try typing
Four pictures of a lighthouse in a storm, in the style of an old poster
A logo for a coffee cart called Morning Owl, simple and round
A 5 second wide clip of a paper boat on a rainy street
A cheerful 30 second song about my dog Biscuit
The sound of rain on a tin roof
Shape, length and how many
Under askr Pro, say them in your words and askr Pro reads them: "vertical", "square" or "wide"; "5 seconds" or "10 seconds"; "three pictures".
With a picture or clip model you chose yourself, an Options chip sits beside the box and sets them for the next one:
- Shape
- Auto, Wide, Tall or Square, for a picture or a clip.
- Length
- Auto, 5 s or 10 s for a clip.
- How many
- Auto, or 1 to 4 pictures.
Left on Auto, the words decide.
Pictures
- Ask for up to four at once: "four pictures of a cabin in the woods".
- Change a picture: attach it with +, Add a photo or file, and say what to change. A picture askr just made works too: "the same, but at night".
- Ask about a photo instead: attach it and ask a question, such as "what plant is this?". askr answers in words.
- Tap a picture to see it full size. Copy copies it, and More, Save downloads it.
Video clips
- Say the shape and the length: vertical, square or wide, 5 or 10 seconds. Each model makes a few set shapes and lengths.
- A clip takes from under a minute to several minutes. A bar shows while it is made. You can leave the page: it lands in the chat and in your Library.
- Start from a picture: attach it and ask for it to move. A picture askr just made works too: ask for "that one, moving". askr Pro picks a model that can start from a picture.
- Tap a clip to play it. Save downloads it, and More, Copy link copies its address.
One video from several pictures
Attach two to ten pictures, such as screenshots of your app, and ask for a video: "make a promo video from these". askr plans the shots from all of them together, makes a clip from each and joins them into one video, in the order you attached them.
- One per picture: each picture starts its own clip, and the clips fade into each other. It works with every model that can start from a picture.
- Transitions: each clip moves from one picture to the next, so the video runs on as one. Kling 3.0, Kling O3 and Veo 3.1 can do this. Under askr Pro, say "with transitions".
- A song: add one and askr makes music the length of the video and lays it under it. Under askr Pro, say "with music".
- Pick a model yourself and the chips under the box set the style, the song and how long each picture's clip is. Under askr Pro, the words do: "a 30 second promo" shares 30 seconds between the pictures.
- The price beside Send is the whole video: every clip, and the song. Credits for all of it are set aside before anything starts. Each clip is charged when it is made, and a clip that cannot be made is left out and not charged.
- Ten pictures take about ten to twenty minutes. The line shows how many clips are done. You can leave the page: the video lands in the chat and in your Library.
Drawn in code
Some pictures and clips come out better written than painted. A strong Claude, GPT or Grok model can write a logo, an icon, a diagram, a chart or a card with words on it as code, and askr turns that into a picture. For a clip it writes an animation, and askr records it as a video: a title, moving words, a chart that builds, a logo coming in. The words come out as real text, far more reliably spelled than in a painted picture, and the lines stay sharp.
- Pick one yourself: tap +, Choose a model, then Image or Video. They are at the end of the list, under Drawn in code or Animated in code, each with the most one can cost.
- Then describe it: the words on it, the colours, the style, and for a clip how it moves. The chip over the box names the model, and your next pictures or clips in that chat are drawn by it too, until you pick another or tap the chip's x to go back to askr Pro.
- Or leave it to askr Pro: when you ask for a logo, an icon, a diagram, a chart or a motion graphic, it can give it to one of these models, and the answer says so: "for a picture drawn in code". At Cheap it starts with Grok Build 0.1, at Balanced with Claude Opus 5.5, and at Performance with Claude Fable 5.1.
- They are not for photos, paintings, 3D scenes, product shots or filmed footage: for those, askr Pro keeps to the picture and video models.
- A clip is 5 or 10 seconds, wide, tall or square, and the chat keeps that shape and length for its next clips and for Try again. It starts from your words, not from a picture.
- A picture takes from a few seconds to a few minutes. A clip takes a minute or more, and lands in the chat when it is ready.
Try typing
A logo for a coffee cart called Morning Owl, simple and round
A bar chart of our sales: January 12, February 18, March 25
A 5 second title card that says Summer Sale, in bold red letters
| Model | What it is | A picture, up to | A clip, up to |
|---|---|---|---|
| Claude Sonnet 5.5 | Claude's quicker model | $0.18 | $0.22 |
| Claude Opus 5.5 | Claude's all-round model | $0.36 | $0.44 |
| Claude Fable 5.1 | Claude's most capable model | $0.91 | $1.11 |
| GPT-6.1 Sol | OpenAI's newest Sol model | $0.27 | $0.34 |
| GPT-6 Sol | The Sol model before 6.1 | $0.27 | $0.34 |
| GPT-6 Astra | OpenAI's most capable model | $1.36 | $1.71 |
| Grok Build 0.1 | SpaceXAI's coding model | $0.06 | $0.07 |
| Grok 4.7 | SpaceXAI's frontier model | $0.17 | $0.21 |
These were the prices on 30 September 2026; Choose a model shows the price at the moment you pick. You pay only for what the model wrote, which is usually well under the most, and a drawing that does not come costs nothing.
Songs, voices and sound effects
- Three kinds: a song or music from a description, a voice reading your words, or a sound effect.
- For a song, say the style, the mood, the tempo and what it sings about. Where the model can make the length you ask for, such as "30 seconds", it does; some song models choose their own length.
- For a voice, ask for your words to be read aloud, and give the words: up to 5,000 characters. askr uses ready-made voices.
- askr Pro reads your words with ElevenLabs v4 Turbo on Cheap and Balanced, and ElevenLabs v4 on Performance, in more than 90 languages. A cue in square brackets, such as [whispers], changes how a line is read.
- For a sound effect, describe the sound: "rain on a tin roof", "a door creaking open".
- A song plays in the chat, and Lyrics shows its words where it has some.
- Every sound is marked Made with AI, and a reading says it is a synthetic voice. Keep that label when you share it. askr offers ready-made voices only: it never copies a real person's voice.
What it costs
Typical prices at each askr Pro level, from the model at the top of that level on 28 September 2026:
| What | Cheap | Balanced | Performance |
|---|---|---|---|
| A picture | $0.04 | $0.08 | $0.16 |
| A 5 second video | $0.07 | $0.40 | $2.30 |
| A 30 second song | $0.08 | $0.10 | $0.60 |
- A bigger picture, a longer clip or a longer song costs more. The strongest clip models cost the most: about $2.30 for 5 seconds at Performance.
- Under askr Pro the price shows under what it made. With a model you chose yourself, the price shows before you send.
- When a clip starts, its price is set aside, and only what it really cost is taken.
- You are never charged for something that failed or was refused.
When a model says no
Each company that makes a model has its own rules. A real, famous person, a brand, or someone else's characters are the usual reasons a picture, a clip or a song is refused. askr Pro rewords the request and tries once more, and says when it did. If that is refused too, it offers a reworded version: tap Yes, make the picture that way, or say it another way in the box. You are not charged for a refusal.
Keeping them
Everything you make is saved in your Library with the words that made it, and you can download any of it there. Something made in a chat is deleted with that chat. Only you can see it, unless you made it in a chat you share with other people.
Still stuck?
A person on the askr team answers you on Telegram. Or ask askr itself.