Together.AI Unveils Enhanced Batch Inference API with Expanded Capabilities

Together.AI has rolled out a major upgrade to its Batch Inference API, delivering a suite of enhancements designed to streamline large-scale AI data processing. The improvements include a revamped user interface, universal model compatibility, and a staggering 3000x increase in rate limits—from 10 million to 30 billion tokens per model per user.

The new UI eliminates cumbersome API calls, offering an intuitive dashboard for batch job creation and monitoring. Universal model support now extends to all serverless models and private deployments, granting users unprecedented flexibility in workload execution. This scalability leap positions the API as a cost-efficient solution for enterprises handling massive datasets, with operational costs reportedly halved.