Google introduces new Gemini text embedding model to developers.
Google introduced a new, experimental addition called Gemini Embedding, an "embedding" model for text, to its Gemini developer API on Friday.
Embedding models convert textual inputs—such as words and phrases—into numerical representations, or embeddings, that encapsulate the semantic meaning of the content. These embeddings are used in various applications, including document retrieval and classification, partly because they can lower costs and improve response speed.
Amazon, Cohere, and OpenAI are among the companies that provide embedding models via their own APIs. While Google has offered embedding models in the past, Gemini Embedding is the first one trained using the Gemini family of AI models.
"Trained on the Gemini model itself, this embedding model has inherited Gemini's grasp of language and nuanced context, making it suitable for a wide array of applications," Google stated in a blog post. "We trained our model to be highly general-purpose, delivering outstanding performance across diverse fields like finance, science, law, search, and more."
According to Google, Gemini Embedding outperforms its previous top-tier model, text-embedding-004, and shows competitive results on widely-used embedding benchmarks. Unlike its predecessor, Gemini Embedding can process larger chunks of text and code in a single input and supports over 100 languages—twice as many as before.
Google notes that Gemini Embedding is currently in an "experimental phase" with limited capacity and is subject to future changes. "We plan to move toward a stable, generally available release in the coming months," the company wrote in its blog post.
Related article
Gemini Spark now manages Google Photos library
Google is expanding its AI integration by enabling Gemini Spark to manage Google Photos libraries. Users can now instruct the agent to edit images, organize albums, generate shared collections from favorite shots, convert concert flyers into calendar
Google rolls out fake call detection to protect against AI deepfake impersonation scams
Google announced on Tuesday that Android is launching fake call detection to protect against AI deepfake impersonation scams. The feature is rolling out globally in Phone by Google to Android 12+ devices this month, starting with Pixel devices.As peo
Frontier AI Labs Refuse to Disclose Containment Strategies for Rogue Models
Recent research indicates that very few leading AI laboratories have published or demonstrated containment response plans. A containment plan defines the procedures for when an AI system attempts to subvert human control, specifying which access righ
Related Special Topic Recommendations
Comments (1)
0/500
Google introduced a new, experimental addition called Gemini Embedding, an "embedding" model for text, to its Gemini developer API on Friday.
Embedding models convert textual inputs—such as words and phrases—into numerical representations, or embeddings, that encapsulate the semantic meaning of the content. These embeddings are used in various applications, including document retrieval and classification, partly because they can lower costs and improve response speed.
Amazon, Cohere, and OpenAI are among the companies that provide embedding models via their own APIs. While Google has offered embedding models in the past, Gemini Embedding is the first one trained using the Gemini family of AI models.
"Trained on the Gemini model itself, this embedding model has inherited Gemini's grasp of language and nuanced context, making it suitable for a wide array of applications," Google stated in a blog post. "We trained our model to be highly general-purpose, delivering outstanding performance across diverse fields like finance, science, law, search, and more."
According to Google, Gemini Embedding outperforms its previous top-tier model, text-embedding-004, and shows competitive results on widely-used embedding benchmarks. Unlike its predecessor, Gemini Embedding can process larger chunks of text and code in a single input and supports over 100 languages—twice as many as before.
Google notes that Gemini Embedding is currently in an "experimental phase" with limited capacity and is subject to future changes. "We plan to move toward a stable, generally available release in the coming months," the company wrote in its blog post.
Gemini Spark now manages Google Photos library
Google is expanding its AI integration by enabling Gemini Spark to manage Google Photos libraries. Users can now instruct the agent to edit images, organize albums, generate shared collections from favorite shots, convert concert flyers into calendar
Google rolls out fake call detection to protect against AI deepfake impersonation scams
Google announced on Tuesday that Android is launching fake call detection to protect against AI deepfake impersonation scams. The feature is rolling out globally in Phone by Google to Android 12+ devices this month, starting with Pixel devices.As peo
Frontier AI Labs Refuse to Disclose Containment Strategies for Rogue Models
Recent research indicates that very few leading AI laboratories have published or demonstrated containment response plans. A containment plan defines the procedures for when an AI system attempts to subvert human control, specifying which access righ





Home






