跑道推出新款AI视频模型,在关键基准测试中胜过谷歌和OpenAI。
Runway rolls out new AI video model that beats Google, OpenAI in key benchmark

原始链接: https://www.cnbc.com/2025/12/01/runway-gen-4-5-video-model-google-open-ai.html

Runway,一家小型人工智能初创公司,发布了Gen 4.5,这是一款新的文本到视频模型,目前在Video Arena排行榜的独立基准测试中表现优于谷歌的Veo 3和OpenAI的Sora 2 Pro。Gen 4.5擅长根据文本提示生成高清视频,展现了对物理、运动和因果关系的深刻理解。 Runway的首席执行官Cristóbal Valenzuela将他们的成功——与拥有万亿美元规模的公司竞争,而团队只有100人——归功于专注的研究和勤奋。该公司成立于2018年,致力于构建用于视频创作的人工智能模型,估值35.5亿美元。 Gen 4.5,绰号“David”,正在向Runway多元化的客户群(包括媒体、工作室和创意人员)推出,并将通过他们的平台、API和合作伙伴提供访问。Runway计划进一步发布重大版本,旨在将人工智能开发民主化,使其超越少数主导企业。

Runway 发布新的 AI 视频模型,在关键基准测试中超越谷歌、OpenAI (cnbc.com) 5 分,tiahura 1 小时前 | 隐藏 | 过去 | 收藏 | 1 条评论 https://runwayml.com/research/introducing-runway-gen-4.5 QuercusMax 4 分钟前 [–] 唉,我们需要的正是更好的假视频。回复 指南 | 常见问题 | 列表 | API | 安全 | 法律 | 申请 YC | 联系 搜索:
相关文章

原文

Mustafa Hatipoglu | Anadolu | Getty Images

Artificial intelligence startup Runway on Monday announced Gen 4.5, a new video model that outperforms similar models from Google and OpenAI in an independent benchmark.

Gen 4.5 allows users to generate high-definition videos based on written prompts that describe the motion and action they want. Runway said the model is good at understanding physics, human motion, camera movements and cause and effect.

The model holds the No. 1 spot on the Video Arena leaderboard, which is maintained by the independent AI benchmarking and analysis company Artificial Analysis. To determine the text-to-video model rankings, people compare two different model outputs and vote for their favorite without knowing which companies are behind them.

Google's Veo 3 model holds second place on the leaderboard, and OpenAI's Sora 2 Pro model is in seventh place.  

"We managed to out-compete trillion-dollar companies with a team of 100 people," Runway CEO Cristóbal Valenzuela told CNBC in an interview. "You can get to frontiers just by being extremely focused and diligent."

Runway was founded in 2018 and earned a spot on CNBC's Disruptor 50 list this year. It conducts AI research and builds video and world models, which are models that are trained on video and observational data to better reflect how the physical world works.

The startup's customers include media organizations, studios, brands, designers, creatives and students. Its valuation has swelled to $3.55 billion, according to PitchBook. Runway's investors include General Atlantic, Baillie Gifford, Nvidia and Salesforce Ventures, among others.

Valenzuela said Gen 4.5 was codenamed "David" in a nod to the biblical story of David and Goliath. The model was "an overnight success that took like seven years," he said. 

"It does feel like a very interesting moment in time where the era of efficiency and research is upon us," Valenzuela said. "[We're] excited to be able to make sure that AI is not monopolized by two or three companies." 

Gen 4.5 is rolling out gradually, but it will be available to all of Runway's customers by the end of the week. Valenzuela said it's the first of several major releases that the company has in store.

"It will be available through Runway's platform, its application programming interface and through some of the company's partners," he said.

WATCH: We tested OpenAI’s Sora 2 AI-video app to find out why Hollywood is worried

联系我们 contact @ memedata.com