Text to video · in ChatGPT and Claude

Text to video AI: turn a prompt or script into a video

Creative Claw turns text into video inside ChatGPT, Claude or Codex: paste a prompt or a script, and it splits it into scenes, makes a frame for each, renders them with Seedance 2.5, Gemini Omni or MiniMax H3 Max, and adds a voiceover and word-timed captions. Our 15-second narrated explainer cost 548 credits ($5.48), with no subscription.

Install in ChatGPT Add to Claude

100 free credits · no subscription · estimate your cost

A narrated explainer from a script

Four scenes in one MiniMax H3 Max generation, narration and captions from a short script. 548 credits ($5.48). See every step

How it works

1

Paste the text

A one-line prompt or a full script. Add the length, format and tone you want.
2

Review the scene plan

Creative Claw splits the script into scenes and makes a start frame for each, so you can fix the look before any video renders.
3

Render the scenes

Up to 15 seconds with several scenes in one MiniMax H3 Max generation, or one clip per scene with Gemini Omni or Seedance 2.5.
4

Add narration and captions

Generate the voiceover from your script (or in your cloned voice) and caption it word by word.

Try this prompt

Use Creative Claw to turn this script into a 15-second 16:9 explainer with one scene per sentence, a calm narration, and captions: "[paste your script]". Show me the scene plan and the credit estimate first.

Open in ChatGPT Open in Claude

Install Creative Claw first so your assistant can run it.

Models and prices

Pay per generation: $10 = 1,000 credits, credits never expire. Full pricing.

Model	Best for	Cost
Gemini Omni Flash	Default: clips with native sound and lip-synced speech, up to 10 s	130 credits / 5 s
Seedance 2.5	Premium fidelity and references; 480p drafts you can finalize at 1080p	119–657 credits / 5 s
MiniMax H3 Max	Up to 15 s with several scenes in one generation	240–480 credits / 15 s
MiniMax H3 Max Fast	Cheapest drafts	40 credits / 5 s
Kling 3.0 Omni	Expressive character motion	140 credits / 5 s
Veo 3.1	Premium Google video	400 credits / 5 s

Estimate your video

Images (start frames, posts)

Voiceover (seconds) Captions (6 credits per minute) Join the shots into one video (10 credits)

Presets:

Estimated total

0 credits · $0.00

Estimates from current list prices; Creative Claw shows the exact cost before each generation. Retakes and edits are charged as new generations. $10 = 1,000 credits.

Cost of a 5-second clip by model

Model	Resolution	Credits per 5 s	Cost
Gemini Omni Flash	720p	130	$1.30
Gemini Omni Flash	1080p	130	$1.30
Seedance 2.5	480p	119	$1.19
Seedance 2.5	720p	267	$2.67
Seedance 2.5	1080p	657	$6.57
MiniMax H3 Max	768p	80	$0.80
MiniMax H3 Max	1080p	160	$1.60
MiniMax H3 Max Fast	768p	40	$0.40
MiniMax H3 Max Fast	1080p	80	$0.80
Seedance 2.0 Mini	480p	39	$0.39
Seedance 2.0 Mini	720p	88	$0.88
Kling 3.0 Omni	1080p	140	$1.40
Veo 3.1	1080p	400	$4.00

Images: Nano Banana 2 16 credits ($0.16), Nano Banana Pro and GPT Image 2.5 30 credits. Voiceover: ElevenLabs v3 is 20 credits per 1,000 characters, about 4 credits for 15 seconds. Full list: pricing.

No monthly plan: compare with Runway, HeyGen and Higgsfield on the pay-as-you-go page.

FAQ

How do I turn text into a video with AI? +

Install Creative Claw in ChatGPT or Claude, paste your prompt or script, and ask for a video with a length and format. It plans the scenes, renders them with a video model, and adds a voiceover and captions.

Can AI turn a script into a narrated video? +

Yes. Creative Claw splits the script into scenes, renders them, and narrates the script with ElevenLabs or Cartesia, including your own cloned voice, with word-timed captions. Our 15-second explainer cost $5.48.

What is the best text-to-video model? +

For several scenes in one clip, MiniMax H3 Max (up to 15 seconds). For the highest fidelity per shot, Seedance 2.5. For everyday clips with sound, Gemini Omni Flash, the default.

How much does text to video cost? +

A 5-second clip costs from 40 credits ($0.40). A 15-second multi-scene clip costs 240 credits ($2.40) at 768p or 480 ($4.80) at 1080p. Narration is about 4 credits per 15 seconds.

Can I make the video in another language? +

Yes. Write the script in that language and generate the narration in it; Cartesia supports 43 languages and ElevenLabs v3 more than 70.

Can I try it for free? +

Yes. New Creative Claw accounts get 100 free credits with no card, enough for two 5-second MiniMax H3 Max Fast clips or six Nano Banana 2 images. After that, $10 buys 1,000 credits that never expire, with no subscription.

Does it work in ChatGPT and Claude? +

Yes. Install Creative Claw from the ChatGPT plugin directory, or add it in Claude at claude.ai/customize/plugins. The same account and credits also work in Codex, Cursor and OpenClaw.

Is text to video different from image to video? +

Text to video starts from words only. Image to video starts from a picture you provide, which gives more control over the look, product or character. Creative Claw does both, and often makes a start frame from your text first.