Streaming and media reliability upgrades

Fixed long-stream handling to reduce truncation and billing misclassification on extended responses.
Improved video-generation reliability across validation and completion flows.
Fixed provider-specific tool-calling and malformed JSON response-format edge cases.
Improved Bedrock dispatch behavior for requested streaming mode.

Get Started

Cut your AI bill, not your usage

Route every request to the right model. Track every dollar you spend. Cut your LLM costs by up to 60%.

Start saving Book a call