Freemium
AI voice generation platform with ultra-realistic speech in 32+ languages
text to speechvoice cloningaudio generation
View detailsKey facts
Pricing
Freemium
Use cases
Developers building video search applications that require the ability to locate specific moments within large video libraries using natural language (verified: 2026-01-29), Content creators needing to analyze and remix video assets by identifying visual and auditory context across their entire media collection (verified: 2026-01-29), Enterprise teams automating video workflows through AI-driven summarization and text generation based on deep video understanding (verified: 2026-01-29)
Strengths
The platform utilizes the Marengo foundation model to analyze temporal relationships between video frames alongside speech and sound for retrieval tasks (verified: 2026-01-29), The Pegasus video-first language model integrates visual and audio information to generate text-based summaries and analysis from video content (verified: 2026-01-29), Developers can access a free tier that includes up to ten hours of video indexing and access to search and embed APIs (verified: 2026-01-29)
Limitations
The free tier limits users to a maximum of 600 minutes of indexing and restricts index access to 90 days (verified: 2026-01-29), Concurrent indexing tasks are capped at five for users on the free plan, which restricts high-volume processing speeds (verified: 2026-01-29)
Last verified
Jan 29, 2026
Plan your next step
Use these links to move from this review into compare and task workflows before committing to a tool stack.
Compare • Browse by task • Guides • Tools • Deals
Priority tasks: Content writing tasks • Code generation tasks • Video generation tasks • Meeting notes tasks • Transcription tasks
Priority guides: AI SEO tools guide • AI coding tools guide • AI video tools guide • AI meeting notes guide
Strengths
- The platform utilizes the Marengo foundation model to analyze temporal relationships between video frames alongside speech and sound for retrieval tasks (verified: 2026-01-29)
- The Pegasus video-first language model integrates visual and audio information to generate text-based summaries and analysis from video content (verified: 2026-01-29)
- Developers can access a free tier that includes up to ten hours of video indexing and access to search and embed APIs (verified: 2026-01-29)
Limitations
- The free tier limits users to a maximum of 600 minutes of indexing and restricts index access to 90 days (verified: 2026-01-29)
- Concurrent indexing tasks are capped at five for users on the free plan, which restricts high-volume processing speeds (verified: 2026-01-29)
FAQ
What specific types of data does the Twelve Labs foundation model analyze within a video file?
The Marengo foundation model analyzes individual video frames and their temporal relationships. It also processes speech and sound data to provide a multi-sensory understanding of the content for search and retrieval tasks (verified: 2026-01-29).
Are there any usage limits for developers who are using the free version of the platform?
Yes, the free tier provides up to 10 hours of indexing with a total limit of 600 minutes. It also restricts each index to 100 videos and limits concurrent indexing tasks to five (verified: 2026-01-29).
How does the Pegasus model differ from the Marengo model in terms of output?
While Marengo focuses on video indexing and retrieval, the Pegasus model is a video-first language model designed to generate text, such as summaries and analysis, by reasoning across visual and audio data (verified: 2026-01-29).
