Skip to content
English

Switch theme

Discover▾

Not sure where to start? This week’s hottest software, models and stacks in three minutes.

AI Apps▾

AI software people use: from assistants like ChatGPT to tools for coding, video and workflows, picked by purpose.

For Agents▾

Software for AI: tools that let your agent (Claude Code, Codex, Cursor…) browse, read docs, run code and remember.

Models▾

Choose models, estimate cost, self-host: prices, rankings, benchmarks, free tiers and compute.

Solutions▾

A goal or an error message? Ready-made stacks, steps and fixes by scenario.

Learn▾

A systematic course from zero to the frontier, with an algorithm library, must-read papers and tutorials.

Frontier▾

Daily AI news, weekly highlights, the latest papers and how AI is changing every discipline.

// ⚡ Inference Acceleration & Serving

Medusa

Adds extra decoding heads so the model drafts its own speculative tokens.

Medusa 多解码头 2024 Advanced extended

Original paper ↓ What to learn first →

IN PLAIN WORDS

No separate draft model: extra heads predict tokens 2, 3, … ahead and tree attention verifies many candidates at once.

Used in: Inference acceleration frameworks

TIMELINE · prerequisites → this → what builds on it

2022 Speculative D…2024 Medusa

01Original paper

Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads

Tianle Cai, Yuhong Li, Zhengyang Geng, et al. · 2024

arXiv ↗ ✓ Verified · arXiv API

02Where it sits

Learn the left column first; the right column builds on this one.

Learn first

Medusa

Then you can learn

  • Nothing in the atlas builds on it yet

Related:EAGLEMulti-Token Prediction (MTP)

Speculative Dec…EAGLEMulti-Token Pre… Medusa

Left: prerequisites · right: builds on this · dashed: related. Full map →

03Learning resources

Official docs, explainers, courses and reference code. All links checked.

Reference code

04Where the course teaches it

U62 Serving Engines PagedAttention, continuous batching, prefix caching, speculative decoding, prefill/decode disaggregation and gateway routing. 8 h7 algorithms4 papers
For AI / data

Data updated: AI algorithms & curriculum(UTC+8)