TensorFlow Lite and MediaPipe have released the experimental MediaPipe LLM Inference API, which allows Large Language Models (LLMs) to run fully on-device across platforms. This new capability streamlines on-device LLM integration for web developers and supports Web, Android, and iOS. The LLM Inference API can be used by

5m read time From developers.googleblog.com
Post cover image
Table of contents
LLM Inference APIModelsModel PerformancePerformance OptimizationsWhat's Next

Sort: