Hedra releases Character-1, a video-focused foundational model

2024-08-02

AI video creation platform Hedra has completed a $10 million seed round of financing, further demonstrating the market's interest in AI investment and generative AI content creation products. Hedra, founded by former employees of NVIDIA, Google, and Meta, has investors including Index Ventures, Abstract, and A16Z Speedrun. During the seed round of financing, Hedra also secretly released its video-focused foundational model, Character-1. The company stated that Character-1, previewed to researchers in June, is the "fastest video foundational model on the market." "So far, creators have faced a trade-off between control and speed when using generative video models," the company said. "Our research preview version of Character-1 in June showed that this trade-off no longer exists. Its speed and ease of use are designed to greatly improve the efficiency of content creators and marketers." Hedra stated that over 350,000 users have used the Character-1 platform and created over 1.6 million videos. The company emphasized that many videos created using Character-1 have gained popularity on Reddit and Instagram. Hedra also highlighted that it allows users to create AI pop stars and AI-generated content, and plans to continue improving its security and content moderation features. The company plans to focus on making its platform more multimodal, in order to "integrate storytelling, sound, and video generation into a unified workflow." Hedra aims to simplify the content creation process and allow users to customize digital avatars and characters using AI. Hedra's Character-1 is the latest video generation model to be released in recent months. Haiper 1.5, secretly released in March, claimed to challenge OpenAI's Sora. RunwayML, one of the pioneers in AI-generated video, publicly launched its Gen-3 Alpha model in July, while Luma AI announced its realistic video platform Dream Machine in June. Another video platform, Captions, completed a $60 million Series C financing in July. In the field of generative AI video creation, major companies such as OpenAI and Google are also involved. OpenAI's Sora has not been publicly released, but the company has collaborated with brands and filmmakers to showcase the platform. Google announced Veo in May.