Subscribe
Topic

Speculative Decoding

Speculative decoding speeds up LLM inference by drafting tokens and verifying them in one pass. Posts explain the math and tree-structured variants, compare Medusa, EAGLE-2 and lookahead decoding, and cover llama.cpp's DFlash 2 and Liquid AI's draft models.

Articles

1-5
  1. llama.cpp Merges DFlash 2 for Up to 2x Faster Decoding

    The open-source llama.cpp project has merged native support for DFlash 2, bringing parallel speculative decoding and substantial inference throughput improvements to local LLM serving across CPU, Apple Silicon, and GPU backends. The implementation, integrated via Pull Request #27342, adds local convolution operators and candidate selector mechanics designed specifically for the DFlash 2 architecture. Non-Autoregressive Speculative Drafting Standard speculative decoding uses a smaller autoreg

    llama.cpp
    4 min read
  2. Speculative Decoding: Mathematical Foundations, Distribution Preservation Proofs, Tree-Structured Verification, and Memory-Bandwidth Amortization

    Autoregressive large language model (LLM) generation suffers from an acute hardware efficiency mismatch during inference. While the prefill phase (processing the input prompt) processes tokens in parallel and achieves high arithmetic intensity on modern matrix accelerators, the decode phase (generating text token-by-token) is fundamentally memory-bandwidth bound. At small batch sizes, each generated token requires transferring the model's entire multi-billion-parameter weight matrix from High-Ba

    AI
    13 min read
  3. Speculative Decoding in Production Serving: Comparing Small Draft Models, Medusa, EAGLE-2, and Lookahead Decoding Architecture, Verification Tree Overhead, and Throughput Economics

    Large language model inference during autoregressive decoding is structurally memory-bandwidth bound. During generation, each forward pass loads the model weight matrices (tens to hundreds of gigabytes) from High-Bandwidth Memory (HBM) into on-chip SRAM to produce a single token. Because the arithmetic intensity is close to zero, modern accelerators like the NVIDIA H100 and B200 spend the vast majority of their compute cycles stalled on memory bus transfers rather than executing matrix multiplic

    AI
    7 min read
  4. Liquid AI Ships LFM2.5-DSpark Draft Models for Up to 3.2x Faster Inference

    Liquid AI has released speculative decoding draft checkpoints for three models across its LFM2.5 series: LFM2.5-1.2B-Instruct, LFM2.5-2.6B, and the mixture-of-experts model LFM2.5-8B-A1B. The release introduces small companion models designed to accelerate auto-regressive generation without altering final token distributions. The draft models are available in Safetensors and GGUF formats on Hugging Face, with immediate support implemented for SGLang and llama.cpp. Architecture and Draft Desig

    AI
    2 min read
  5. Tree-Structured Speculative Decoding: How Multi-Candidate Trees and Tree Attention Accelerate LLM Serving

    Tree-Structured Speculative Decoding: How Multi-Candidate Trees and Tree Attention Accelerate LLM Serving Large language model inference is fundamentally constrained by memory bandwidth during the auto-regressive decoding phase. Because each token generation step requires loading billions of model parameters from high-bandwidth memory (HBM) to compute units for a single token, standard auto-regressive generation operates at low arithmetic intensity. Speculative decoding addresses this bottlene

    Speculative Decoding
    7 min read