Blog

Optimizing Krea 3 on AMD Instinct MI355X with AI researchers

Numerical precision, GPU kernels and communication choices behind Krea 3’s image-editing inference.

The Hedra Developer API: models, SDKs, MCP and the console in one place

The October 2026 developer-platform announcement. For current setup and model support, use the documentation.

Seedance, Kling, Veo and MiniMax API Pricing: What 10 Seconds of AI Video Costs

One fixed job, a 10-second 1080p clip with native audio from a start frame, priced across models at the time of writing.

Hedra's MCP and CLI: Bring Hedra inference into your own agents

The August 2026 introduction to Hedra’s MCP and CLI. Current tools, authentication and billing are documented separately.

Texas Rangers Revolutionize Fan Engagement with Hedra

Hedra for Developers: Our models, now available through API, SDK, CLI & MCP

Our launch announcement for developer access, as of its publication date. Current setup steps are in the documentation.

Benchmarking a Multimodal Agent

Michael Loh on evaluating Play through repeatable multi-turn scenarios.

Hedra Omnia: A new unified base model for post-training

Michael Lingelbach on Omnia’s direction as a unified base model for partner post-training.