The Hugging Face Inference API makes it easy to integrate any public or private AI model with Golioth Pipelines for use-cases such as text, video, and audio analysis.
Golioth enables AI applications by pushing TF Lite model updates directly to devices. These can be stored locally and begin running without a device restart. This example showcases the Golioth SDK running on ESP-IDF.
OpenAI's Structured Outputs allow callers to provide a JSON schema to define the structure in which responses should be formatted. When combined with Golioth Pipelines, this enables converting device data from an unknown to a known structure.
Today, we are thrilled to announce the launch of Golioth for AI, a comprehensive set of features designed to simplify and enhance the integration of AI into IoT products.
Golioth is open sourcing the PCBs in their hardware platform ("Aludel"). This includes a main processing and communication board ("Elixir") and an ePaper, touch enabled PCB front panel board ("Ostentus")
