Pinned Loading

1. MoR MoR Public

   PyTorch implementation of "Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation" (NeurIPS 2025, Google DeepMind)

   Python

2. LLM-Counsel LLM-Counsel Public

   LLM Counsel is a FastAPI service that routes queries to one or more language models, aggregates responses, and returns a final answer with confidence and metrics.

   Python

3. RRT RRT Public

   JAX/Flax implementation of "Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA" (Bae et al., ICLR 2025, Google DeepMind)

   Python 1