# Multi-Token Prediction Heads and Speculative Decoding > Source-anchored research for Multi-Token Prediction Heads and Speculative Decoding. Run status: COMPLETED. Parent: [Local Inference Acceleration: Memory Architectures, Metal Kernels, and Latency Optimization](https://llms-explorer.com/tree/local-inference-acceleration-and-kernels/) · 2 facets · 2 facts · page: https://llms-explorer.com/tree/multi-token-prediction-heads-and-speculative-decoding/ ## Measurements and reference values - On memory-bandwidth-bound local hardware like Apple Silicon M5 Max, speculative decoding and MTP boost effective generation throughput by 20% to 40% without modifying output token probabilities. (high confidence) — [source](https://huggingface.co/Qwen) ## Facts and statements - Multi-Token Prediction (MTP) generates candidate future tokens in parallel and verifies them in a single transformer forward pass, amortizing weight memory reads across multiple accepted tokens. (high confidence) — [source](https://huggingface.co/Qwen)