Speculative Decoding Algorithms LLM Tools

Implementations, frameworks, and optimization techniques for speculative decoding that accelerate LLM inference through draft model speculation and verification. Does NOT include general LLM inference optimization, quantization methods, or non-speculative decoding strategies.

There are 9 speculative decoding algorithms tools tracked. The highest-rated is vitali87/speculant-graph at 35/100 with 9 stars.

Get all 9 projects as JSON

curl "https://pt-edge.onrender.com/api/v1/datasets/quality?domain=llm-tools&subcategory=speculative-decoding-algorithms&limit=20"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.

#	Tool	Score	Tier	Stars	Language
1	vitali87/speculant-graph Graph drafts, LLM verifies: a novel speculative decoding framework	35	Emerging	9	Python
2	hsj576/GRIFFIN Official Implementation of "GRIFFIN: Effective Token Alignment for Faster...	26	Experimental	18	Python
3	Hambaobao/HCP-Coder Hierarchical Context Pruning (HCP): A strategy to optimize real-world code...	24	Experimental	16	Python
4	levvius/adaptive-speculative-decoding Adaptive speculative decoding for LLM inference latency optimization	22	Experimental	—	Python
5	hsj576/GTO Official Implementation of "Bridging Draft Policy Misalignment: Group Tree...	22	Experimental	3	Python
6	CyberCoder-IITM/HaloSpec Adaptive speculative decoding benchmark with runtime perturbation and...	19	Experimental	—	Rust
7	Geralt-Targaryen/Awesome-Speculative-Decoding Reading notes on Speculative Decoding papers	18	Experimental	25	—
8	OdedMous/DP-Decoding-in-LLM Experiment a differentially private decoding strategy for LLMs.	15	Experimental	2	HTML
9	Hassan-Sarwat/efficient-speculative-decoding Improving both reasoning speed of LLM using Chain of Draft fine tuning and...	11	Experimental	—	Jupyter Notebook