GLiClass is an open-source zero-shot sequence classification model inspired by GLiNER that achieves comparable performance to cross-encoder models while being 10 times faster through single forward pass classification. It supports hierarchical labels, in-context examples, custom prompts, and long document chunking for improved accuracy and flexibility.
Show-Harness is an embodied interface that enables vision-language models to control robots through discrete semantic actions, allowing both closed-source frontier VLMs and fine-tuned open-source models to perform zero-shot and low-cost robot control. The system includes GUMI, a GUI manipulation interface for demonstration collection, and demonstrates robust generalization across tasks, embodiments, and environments without requiring additional model capacity or embodiment-specific pretraining.