FasterDecoding

community

https://github.com/FasterDecoding

Activity Feed Request to join this org

AI & ML interests

Making model inference more efficient by model-system codesign.

Recent Activity

Gsunshine authored a paper about 1 month ago

Representation Fréchet Loss for Visual Generation

tianlecai authored a paper 10 months ago

FutureX: An Advanced Live Benchmark for LLM Agents in Future Prediction

tianlecai authored a paper about 2 years ago

SnapKV: LLM Knows What You are Looking for Before Generation

View all activity

authored a paper about 1 month ago

Representation Fréchet Loss for Visual Generation

Paper • 2604.28190 • Published Apr 30 • 32

authored a paper 10 months ago

FutureX: An Advanced Live Benchmark for LLM Agents in Future Prediction

Paper • 2508.11987 • Published Aug 16, 2025 • 73

authored 2 papers about 2 years ago

SnapKV: LLM Knows What You are Looking for Before Generation

Paper • 2404.14469 • Published Apr 22, 2024 • 27

JetMoE: Reaching Llama2 Performance with 0.1M Dollars

Paper • 2404.07413 • Published Apr 11, 2024 • 38

authored a paper over 2 years ago

One-Step Diffusion Distillation via Deep Equilibrium Models

Paper • 2401.08639 • Published Dec 12, 2023

authored 2 papers over 2 years ago

DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models

Paper • 2402.19481 • Published Feb 29, 2024 • 22

BitDelta: Your Fine-Tune May Only Be Worth One Bit

Paper • 2402.10193 • Published Feb 15, 2024 • 21

authored a paper over 2 years ago

BitDelta: Your Fine-Tune May Only Be Worth One Bit

Paper • 2402.10193 • Published Feb 15, 2024 • 21

updated 2 models over 2 years ago

FasterDecoding/BitDelta_Mistral_combo

Updated Feb 14, 2024

FasterDecoding/medusa-1.0-vicuna-13b-v1.5

Text Generation • Updated Jan 25, 2024 • 10 • 1

authored a paper over 2 years ago

Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads

Paper • 2401.10774 • Published Jan 19, 2024 • 60

authored a paper over 2 years ago

Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads

Paper • 2401.10774 • Published Jan 19, 2024 • 60

updated 3 models over 2 years ago

FasterDecoding/medusa-1.0-vicuna-33b-v1.3

Text Generation • Updated Dec 18, 2023 • 1

FasterDecoding/medusa-1.0-zephyr-7b-beta

Text Generation • Updated Dec 18, 2023 • 411 • 1

FasterDecoding/medusa-v1.0-vicuna-7b-v1.5

Text Generation • Updated Oct 29, 2023 • 506

updated 3 models almost 3 years ago

FasterDecoding/medusa-vicuna-33b-v1.3

Updated Sep 11, 2023 • 11 • 4

FasterDecoding/medusa-vicuna-13b-v1.3

Updated Sep 11, 2023 • 12 • 5

FasterDecoding/medusa-vicuna-7b-v1.3

Updated Sep 11, 2023 • 277 • 17

updated a Space almost 3 years ago

README

authored a paper about 3 years ago

Large Language Models as Tool Makers

Paper • 2305.17126 • Published May 26, 2023 • 4