Baseten Expands Decision Models With 128K Context, Image Input
Baseten has extended the capabilities of its decision models, adding support for a 128,000-token context window along with image input processing, according to Unite.AI.
The expanded context length allows the models to process significantly larger volumes of text in a single request, a capability that can benefit use cases requiring analysis of lengthy documents or extended conversational histories. The addition of image input broadens the models beyond text-only processing, enabling multimodal inputs for decision-making tasks.
Baseten operates in the infrastructure layer of the AI industry, providing platforms for deploying and serving machine learning models. Decision models, as a category, are typically used to support automated or semi-automated choices in business and technical workflows, drawing on structured and unstructured data to produce recommendations or outcomes.
Unite.AI did not provide further technical specifics in the available source material regarding benchmark performance, pricing changes, or specific customer use cases tied to the update. The report frames the expansion as part of a broader trend among AI infrastructure providers to increase the flexibility and scale of the models they support, as enterprises increasingly look for tools capable of handling diverse and larger inputs.
Longer context windows have become a focal point of competition across the AI tooling landscape, with vendors racing to support larger inputs without sacrificing speed or accuracy. Similarly, multimodal input support, combining text and images, has become a common feature add as developers build more complex applications that require reasoning across different data types.
Further details on the rollout timeline or availability of the updated decision models were not specified in the source report.
Based on reporting by www.unite.ai.
