Footer

Your go-to destination for discovering trending open-source projects and uncovering the insights that matter

Legal

  • Terms and conditions
  • Privacy policy

More

  • Pricing
  • Advertise
  • onedump
  • gosnakego
  • Contact us

Social

  • X (Twitter)
  • GitHub
Trendshift owl
What they build speaks louder than words
© 2026 Trendshift, created by Julian Li with ❤️
SignalAdvertise
Trendshift
SignalAdvertise
Sign in
Trendshift owl

Discover trends that matter

DailyWeeklyMonthlyYearly
Live mentionsTopics
RepositoriesDevelopers
InsightsStats
Bifrost
Bifrost
ClawBench
ClawBench

vllm-project/speculators

A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM

Visit GitHub
Python
920
282
34 contributors
Apache License 2.0
website
FeaturedAdvertise to 180K+ monthly visitors
Bifrost

Enterprise AI Gateway Built For Scale: Route, Govern, and Secure All Your AI Traffic

ClawBench

a benchmark for personal agents

Momentum

How fast vllm-project/speculators is gaining stars, and where it lands on Trendshift and GitHub Trending

Activity and trending

Stars, forks and merged pull requests each day or month, lined up with the repository's rank on Trendshift Trending

Aug 25 – Oct 24, 2026

Stars +160 in 61 days

051015
14

Forks +74 in 61 days

0246
6

Merged PRs 138 in 61 days

051015
11

Trendshift Trendingwhere the repository ranked each day it made the top 25

Not in the Trendshift top 25 in this period

Aug 25Sep 2Sep 10Sep 18Sep 26Oct 4Oct 12Oct 20
The left and right arrow keys step from day to day; Home and End go to the first and the last

Stars, forks, merged pull requests and Trendshift Trending rank, by day
DayStars gainedForks gainedMerged PRs
Tue, Aug 25, 2026+3+33
Wed, Aug 26, 2026+3+15
Thu, Aug 27, 2026+2+17
Fri, Aug 28, 2026+9+00
Sat, Aug 29, 2026+5+10
Sun, Aug 30, 2026+1+21
Mon, Aug 31, 2026+4+08
Tue, Sep 1, 2026+4+25
Wed, Sep 2, 2026+4+07
Thu, Sep 3, 2026+8+13
Fri, Sep 4, 2026+3+22
Sat, Sep 5, 2026+1+10
Sun, Sep 6, 2026+0+11
Mon, Sep 7, 2026+6+10
Tue, Sep 8, 2026+8+65
Wed, Sep 9, 2026+2+13
Thu, Sep 10, 2026+1+12
Fri, Sep 11, 2026+1+02
Sat, Sep 12, 2026+2+00
Sun, Sep 13, 2026+0+00
Mon, Sep 14, 2026+4+41
Tue, Sep 15, 2026+2+04
Wed, Sep 16, 2026+2+03
Thu, Sep 17, 2026+4+12
Fri, Sep 18, 2026+3+32
Sat, Sep 19, 2026+1+11
Sun, Sep 20, 2026+5+20
Mon, Sep 21, 2026+0+24
Tue, Sep 22, 2026+1+05
Wed, Sep 23, 2026+1+01
Thu, Sep 24, 2026+2+00
Fri, Sep 25, 2026+0+22
Sat, Sep 26, 2026+1+00
Sun, Sep 27, 2026+1+02
Mon, Sep 28, 2026+2+30
Tue, Sep 29, 2026+2+14
Wed, Sep 30, 2026+1+13
Thu, Oct 1, 2026+0+01
Fri, Oct 2, 2026+1+011
Sat, Oct 3, 2026+0+00
Sun, Oct 4, 2026+1+10
Mon, Oct 5, 2026+1+13
Tue, Oct 6, 2026+1+40
Wed, Oct 7, 2026+1+12
Thu, Oct 8, 2026+1+14
Fri, Oct 9, 2026+0+22
Sat, Oct 10, 2026+5+00
Sun, Oct 11, 2026+2+20
Mon, Oct 12, 2026+5+23
Tue, Oct 13, 2026+4+21
Wed, Oct 14, 2026+2+13
Thu, Oct 15, 2026+2+23
Fri, Oct 16, 2026+3+12
Sat, Oct 17, 2026+0+20
Sun, Oct 18, 2026+0+10
Mon, Oct 19, 2026+14+09
Tue, Oct 20, 2026+8+31
Wed, Oct 21, 2026+4+24
Thu, Oct 22, 2026+3+21
Fri, Oct 23, 2026+1+00
Sat, Oct 24, 2026+2+00
Total+160+74138
  • 1#1 on the list
  • 7Rank that day
  • Not trending

Social mentions

Recent discussions about this repository across the web
  1. Oct 141 post

    • Join us for a chat about speculative decoding! We'll share some practical tips and tricks for both training and deployment in production.

      Read full on X, posted by @_EldarKurtic

      @_EldarKurtic · x.com

  2. 27 days earlier
  3. Sep 171 post

    • 🚀 D-Spark for GLM-5.3 beats native MTP on 8×B300: 29% faster single-stream decoding and 16% higher peak throughput. On MRCR, acceptance holds through 1M context, averaging 4.293 accepted tokens in…

      Read full on X, posted by @RedHat_AI

      @RedHat_AI · x.com

  4. 2 months earlier
  5. Jul 91 post

    • The primary difference / benefit of this checkpoint over other NVFP4 variants is that the attention layers have also been optimally quantized on a per block basis to FP8 Combining this checkpoint…

      Read full on X, posted by @dsikka84

      @dsikka84 · x.com

  6. 1 month earlier
  7. May 301 post

    • Laguna XS.2 from @poolsideai is a 33B MoE built for agentic coding. Red Hat AI trained a DFlash speculator for it: 0.6B drafter, 8 tokens per pass, no quality loss. FP8, NVFP4, and INT4 checkpoints…

      Read full on X, posted by @RedHat_AI

      @RedHat_AI · x.com

You might also like

Repositories close to vllm-project/speculators
FeaturedAdvertise to 180K+ monthly visitors
Bifrost

Enterprise AI Gateway Built For Scale: Route, Govern, and Secure All Your AI Traffic

ClawBench

a benchmark for personal agents

lightseekorg/TorchSpec

Python

A PyTorch native library for training speculative decoding models

26669

deepseek-ai/DeepSpec

Python

DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms

7.2k713

sgl-project/SpecForge

Python

Train speculative decoding models effortlessly and port them smoothly to SGLang serving.

1.2k373

tanishqkumar/ssd

Python

A lightweight inference engine supporting speculative speculative decoding (SSD).

1k82

hao-ai-lab/JetSpec

Python

JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Causal Parallel Tree Drafting

24116

SafeAILab/EAGLE

Python

Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).

2.6k294

ifm-ai/uno

Python

Unlocking Lossless Speedups in LLMs via Discrete Diffusion

10717

z-lab/dflash

Python

DFlash: Block Diffusion for Flash Speculative Decoding

6.2k475

apple-aiml-research/ml-recurrent-drafter

Python
22519