Running 119 Unlocking On-Policy Distillation for Any Model Family 📝 119 Explore on-policy distillation visualization for any model
SpecBundle Collection A collection of production-grade draft models for speculative decoding • 28 items • Updated about 3 hours ago • 19
Running Agents 1.51k Big Code Models Leaderboard 📈 1.51k Explore code model leaderboard and submit evaluations
Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads Paper • 2401.10774 • Published Jan 19, 2024 • 60
Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models Paper • 2312.06585 • Published Dec 11, 2023 • 29