RelayLLM: Efficient Reasoning via Collaborative Decoding
Open paper
Benchmark
Model
Rank
Results
mathematical-reasoning-on-minerva-math
Qwen3-0.6B (RelayLLM Difficulty-Aware-Reward)
#39
Accuracy: 23.53
multi-task-language-understanding-on-mmlu-pro
Qwen3-1.7B (RelayLLM Difficulty-Aware-Reward)
#98
Accuracy: 59.03
Rank counts only results with a code link.