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

SignalAdvertise
DailyWeeklyMonthlyYearly
Live mentionsTopics
RepositoriesDevelopers
InsightsStats

Bifrost
Bifrost
Busbar
Busbar
Trina AX
Trina AX
OfinTech Desk
OfinTech Desk
Omnigraph
Omnigraph
It's a plan
It's a plan

FlashML-org/FreeToken

FreeToken brings datacenter-scale model serving to your desktop. Run massive models locally, fast and efficiently.

#LLM inference
Visit GitHub
Python
17.6k
1.8k
3 contributors
Apache License 2.0
website
AchievementsView all →
FeaturedAdvertise to 180K+ monthly visitors
Bifrost

Fastest Enterprise AI Gateway for Routing, Governing, and Securing all your AI Traffic

Busbar

Control what AI can do before it acts

Trina AX

A self-hosted team AI workspace for company files, with per-person folder permissions

OfinTech Desk

An open-source AI workspace for finance and accounting teams, in which every figure is traced to its source

Omnigraph

Versioned graph database on object storage. Git-style workflows and multimodal retrieval

It's a plan

Open-source Linear alternative

An issue tracker where AI agents take the work you assign them

Momentum

How fast FlashML-org/FreeToken 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

Sep 23 – Nov 22, 2026

Stars +4,000 in 61 days

05001K
654

Forks +462 in 61 days

02040
36

Merged PRs 157 in 61 days

051015
12

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

Top 25 in Python

9711
Sep 23Oct 1Oct 9Oct 17Oct 25Nov 2Nov 10Nov 18
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 PRsPython
Wed, Sep 23, 2026+89+71Not trending
Thu, Sep 24, 2026+53+111Not trending
Fri, Sep 25, 2026+50+111Not trending
Sat, Sep 26, 2026+44+41Not trending
Sun, Sep 27, 2026+59+80Not trending
Mon, Sep 28, 2026+35+40Not trending
Tue, Sep 29, 2026+42+131Not trending
Wed, Sep 30, 2026+46+91Not trending
Thu, Oct 1, 2026+48+52Not trending
Fri, Oct 2, 2026+38+80Not trending
Sat, Oct 3, 2026+22+40Not trending
Sun, Oct 4, 2026+36+61Not trending
Mon, Oct 5, 2026+31+93Not trending
Tue, Oct 6, 2026+31+60Not trending
Wed, Oct 7, 2026+19+24Not trending
Thu, Oct 8, 2026+35+26Not trending
Fri, Oct 9, 2026+39+43Not trending
Sat, Oct 10, 2026+34+33Not trending
Sun, Oct 11, 2026+40+31Not trending
Mon, Oct 12, 2026+39+61Not trending
Tue, Oct 13, 2026+36+31Not trending
Wed, Oct 14, 2026+19+62Not trending
Thu, Oct 15, 2026+25+83Not trending
Fri, Oct 16, 2026+20+812Not trending
Sat, Oct 17, 2026+20+49Not trending
Sun, Oct 18, 2026+20+13Not trending
Mon, Oct 19, 2026+19+15Not trending
Tue, Oct 20, 2026+50+97Not trending
Wed, Oct 21, 2026+37+45Not trending
Thu, Oct 22, 2026+21+95Not trending
Fri, Oct 23, 2026+19+411Not trending
Sat, Oct 24, 2026+22+23Not trending
Sun, Oct 25, 2026+21+31Not trending
Mon, Oct 26, 2026+12+52Not trending
Tue, Oct 27, 2026+12+15Not trending
Wed, Oct 28, 2026+20+41Not trending
Thu, Oct 29, 2026+13+45Not trending
Fri, Oct 30, 2026+14+21Not trending
Sat, Oct 31, 2026+144+140Not trending
Sun, Nov 1, 2026+392+320#9
Mon, Nov 2, 2026+654+343#7
Tue, Nov 3, 2026+470+364#11
Wed, Nov 4, 2026+112+116Not trending
Thu, Nov 5, 2026+99+144Not trending
Fri, Nov 6, 2026+57+102Not trending
Sat, Nov 7, 2026+44+63Not trending
Sun, Nov 8, 2026+25+62Not trending
Mon, Nov 9, 2026+90+152Not trending
Tue, Nov 10, 2026+66+123Not trending
Wed, Nov 11, 2026+46+23Not trending
Thu, Nov 12, 2026+49+50Not trending
Fri, Nov 13, 2026+74+83Not trending
Sat, Nov 14, 2026+28+53Not trending
Sun, Nov 15, 2026+32+53Not trending
Mon, Nov 16, 2026+26+51Not trending
Tue, Nov 17, 2026+25+50Not trending
Wed, Nov 18, 2026+20+21Not trending
Thu, Nov 19, 2026+24+31Not trending
Fri, Nov 20, 2026+95+121Not trending
Sat, Nov 21, 2026+162+150Not trending
Sun, Nov 22, 2026+36+21Not trending
Total+4,000+462157
  • 1#1 on the list
  • 7Rank that day
  • Not trending
2
  • GitHub trending history

    Where the repository ranked on GitHub Trending, one dot for each day it was on a list

    Nov 1 – Nov 2, 2026Best #6

    #1#5#10#15#20#25
    Top 10
    Nov 1, 2026Nov 4, 2026
    The left and right arrow keys step from day to day; Home and End go to the first and the last

    Mentions and achievements

    Where people are talking about FlashML-org/FreeToken, and every badge it has earned
    1. Nov 201 post
      • 本地跑大模型最烦显存不够。FreeToken 思路对了:MoE 模型不整个塞进显卡,每个 token 只用一小撮专家,冷专家扔内存、用谁调谁。伯克利和 MIT 的团队搞的,1.7 万星。可我扒仓库犯嘀咕:一个跑分都不给,还挂着性能回退。你信宣传还是信实测? #AI #OpenSource #LocalLLM

        Read full on X, posted by @FGallagher45519

        @FGallagher45519 · x.com

    2. Nov 151 post
      • 伯克利、MIT等搞了一个开源推理引擎:FreeToken,让个人电脑也能运行数千亿参数的MoE,RTX 5090实测DeepSeek 284B达到22–25tok/s 在消费级硬件测试中,解码吞吐量比现有本地推理引擎快约1.3-2.1倍 能接入Claude Code等Coding Agent,让本地开源模型做它们的推理后端…

        Read full on X, posted by @aigclink

        @aigclink · x.com

    3. Nov 101 post
      • From UC Berkeley researchers: serve giant MoE models on consumer gaming GPUs. Your gaming rig could become an AI inference server. Cheaper compute, more profit. 🎮

        Read full on X, posted by @BRKoshiya

        @BRKoshiya · x.com

    4. Nov 91 post
      • UC Berkeley open-sourced FreeToken! FreeToken is an edge-native inference engine for running frontier-scale MoE models on consumer hardware. GPU, CPU, host memory, and PCIe interconnects treated as…

        Read full on X, posted by @Sumanth_077

        @Sumanth_077 · x.com

    5. Nov 81 post
      • FreeToken 这次做的几个优化挺实在,核心就是一件事:让大模型在有限显存下跑得更快,减少数据搬运和重复计算。 主要做了 4 个优化: 1. GPU 缓存常用专家。 不需要把所有专家都塞进显存,常用的留在 GPU,不常用的放在内存里。需要时再决定搬到 GPU,还是直接交给 CPU 计算。 2. 数据搬运和计算同时进行。 GPU…

        Read full on X, posted by @sitinme

        @sitinme · x.com

    6. Nov 42 posts
      • Your gaming PC can now run a 284B AI model. Berkeley and MIT researchers open-sourced FreeToken, a free engine for running large AI models on your own graphics card. their paper shows: • a 35B model…

        Read full on X, posted by @simplifyinAI

        @simplifyinAI · x.com

      • spec-kit (about 146k stars): templates and commands from GitHub that have your coding agent work from a written spec and plan first. Handy if your agent keeps drifting off the brief.

        Read full on X, posted by @CreativeAiNinja

        @CreativeAiNinja · x.com

    7. Nov 33 posts
      • Top AI repos on GitHub today | Wed, Nov 4 • vercel-labs/agent-desktop: A command line tool that lets an AI agent use your computer. It can click around and read what is on screen. •…

        Read full on X, posted by @VT_zero1

        @VT_zero1 · x.com

      • FreeToken is an open-source engine for running big Mixture-of-Experts models on the gaming PC you already own. It splits work between GPU and CPU, caches experts and supports RTX 30/40/50 cards.…

        Read full on X, posted by @SimonObserves

        @SimonObserves · x.com

      • if you're a vibecoder burning money on API tokens, you might wanna see this. it's called FreeToken. it's an open source engine that runs big open-weight models (the mixture-of-experts kind, like…

        Read full on X, posted by @mayorxbt

        @mayorxbt · x.com

    You might also like

    Repositories close to FlashML-org/FreeToken
    FeaturedAdvertise to 180K+ monthly visitors
    It's a plan

    Open-source Linear alternative

    An issue tracker where AI agents take the work you assign them

    Omnigraph

    Versioned graph database on object storage. Git-style workflows and multimodal retrieval

    OfinTech Desk

    An open-source AI workspace for finance and accounting teams, in which every figure is traced to its source

    Trina AX

    A self-hosted team AI workspace for company files, with per-person folder permissions

    Bifrost

    Fastest Enterprise AI Gateway for Routing, Governing, and Securing all your AI Traffic

    Busbar

    Control what AI can do before it acts

    Tiiny-AI/PowerInfer

    C++

    High-speed Large Language Model Serving for Local Deployment

    9.9k668

    bigscience-workshop/petals

    Python

    🌸 Run LLMs at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading

    10.7k659

    eddoursul/Lamina

    C++

    A custom inference engine

    675

    TandemLLM/TandemEngine

    Python

    TandemEngine: local inference that drafts and verifies its own tokens, behind an OpenAI-compatible API.

    8318

    nomic-ai/gpt4all

    C++

    GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.

    77.3k8.4k

    runmyai/run-my-ai

    Dart

    The best local AI experience. Powered by TokenWing.

    754

    jitllm/jitllm

    Go

    Smart inference engine that adapts to your hardware to run models faster, even when they don’t fit in memory

    586

    runmyai/tokenwing

    C++

    Run huge AI models on your own PC, even ones bigger than your memory.

    947

    mlc-ai/mlc-llm

    Python

    Universal LLM Deployment Engine with ML Compilation

    23.5k2.4k