Stability AI is a privately owned artificial intelligence company founded in 2019 and headquartered in the United Kingdom. Employing approximately 170 people, the company develops open-source generative AI models across various media, including image, language, audio, video, 3D, and biological data. In 2023, the company reported revenues of approximately $10 million.

Revenue

Founded

2019

Headcount

170

Headquarters

UK

Primary Segment

Generative AI

Ownership

Privately Owned

Deployments

1

News Summary:

On an unspecified date, Stability AI announced the availability of its Image Services in Amazon Bedrock, offering media editing capabilities via the Amazon Bedrock API. These services build upon existing Stable Diffusion 3.5 models (SD3.5), Stable Image Core, and Ultra models already integrated into Amazon Bedrock. On another unspecified date, the company further detailed the integration, highlighting nine tools designed to improve image creation and modification for businesses. These tools provide precise control over image creation and editing, extending the capabilities of Stable Diffusion and Stable Image models. Also on an unspecified date, a partnership expansion with Amazon Web Services was announced, bringing Image Services—including tools like Inpaint Erase and Remove Background—to Amazon Bedrock as managed API services for easy integration into applications. A separate news story, unrelated to the company's products or services, discussed the legal challenges surrounding generative AI and training data in the UK, referencing a Getty Images lawsuit. Finally, on an unspecified date, Emad Mostaque, CEO of Stability AI, commented on the potential disruptive impact of AI on economic systems.

Similar Companies

Subscribe for full access to Stability AI's profile
Stability AI offers 3 products in the enterprise tech industry. Stability AI's product portfolio comprises of generative AI.
Subscribe for full access to Stability AI's products in full detail
Stability AI's revenues were less than $50M in 2023. All the revenue comes from aI content generation.
Subscribe for full access to Stability AI's revenue in full detail
Dec
13th
2025
18:30
Artificial Intelligence, B2B Tag
音乐生成模型综述:从符号作曲到音频域大模型、评测体系与产业化趋势
[Translated] 目录 1 引言 2 音乐生成任务的表示与问题设定:谱面世界与声学世界如何握手 表1 生成任务的主流表征与适用场景 3 符号音乐生成:把“乐谱当语言”的漫长道路 表2 符号域代表性工作与关键思想 4 从“生成音频”到“生成可压缩 token”:神经音频编解码器与离散化 表3 音频离散化/表示学习的关键支撑工作 5 音频域大模型:自回归、掩码并行、扩散/流匹配三大范式 表4 音频域生成范式对比:速度、质量、可控性的代价结构 6 可控生成与对齐:从“能生成”到“能听懂你的话” 表5 可控性“接口化”的常见手段:你能控制什么,取决于你把什么变成条件 7 评测基准与评估体系:为什么“听起来不错”无法成为科学指标 表6 常见评测指标/方法与它们的“盲点” 8 数据集与训练数据:从“公开基准”到“版权与溯源”成为第一等约束 9 统一化趋势:从单任务模型到“音频基础模型”与多模态创作链 10 结语:未来几年真正决定路线胜负的,可能是评测与合规而非模型规模 参考文献与资料(精选) “让机器写音乐”这件事,过去很长时间都像是在两种世界之间搭桥:一边是乐理、谱面、结构、动机与发展这些高度抽象的符号体系;另一边是声学、音色、混音、空间感与“像真的一样”的听感。
Chinese Software Developer Network
Stability AI has customers like Sky News.

Example Customers

Subscribe for full access to Stability AI's profile