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
ClawBench
ClawBench
Bifrost
Bifrost
Busbar
Busbar

hr98w/jev-visual

An educational Jev-like visual inference experiment on Apple Silicon: shared context, direct candidate scoring, and local visual demos.

Visit GitHub
Python
322
29
1 contributors
MIT License
FeaturedAdvertise to 180K+ monthly visitors
ClawBench

a benchmark for personal agents

Bifrost

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

Busbar

Control what AI can do before it acts

Momentum

How fast hr98w/jev-visual 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 16 – Oct 22, 2026

Stars +322 in 37 days

050100150
103

Forks +29 in 37 days

051015
11

Merged PRs 0 in 37 days

01

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

Not in the Trendshift top 25 in this period

Sep 16Sep 21Sep 26Oct 1Oct 6Oct 11Oct 16Oct 21
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
Wed, Sep 16, 2026+0+00
Thu, Sep 17, 2026+5+10
Fri, Sep 18, 2026+103+110
Sat, Sep 19, 2026+36+80
Sun, Sep 20, 2026+24+30
Mon, Sep 21, 2026+47+10
Tue, Sep 22, 2026+24+10
Wed, Sep 23, 2026+20+00
Thu, Sep 24, 2026+19+10
Fri, Sep 25, 2026+5+00
Sat, Sep 26, 2026+3+00
Sun, Sep 27, 2026+2+00
Mon, Sep 28, 2026+7+00
Tue, Sep 29, 2026+3+00
Wed, Sep 30, 2026+4+10
Thu, Oct 1, 2026+0+00
Fri, Oct 2, 2026+0+00
Sat, Oct 3, 2026+1+00
Sun, Oct 4, 2026+0+10
Mon, Oct 5, 2026+3+00
Tue, Oct 6, 2026+3+10
Wed, Oct 7, 2026+2+00
Thu, Oct 8, 2026+2+00
Fri, Oct 9, 2026+2+00
Sat, Oct 10, 2026+0+00
Sun, Oct 11, 2026+1+00
Mon, Oct 12, 2026+0+00
Tue, Oct 13, 2026+0+00
Wed, Oct 14, 2026+1+00
Thu, Oct 15, 2026+0+00
Fri, Oct 16, 2026+0+00
Sat, Oct 17, 2026+1+00
Sun, Oct 18, 2026+0+00
Mon, Oct 19, 2026+3+00
Tue, Oct 20, 2026+0+00
Wed, Oct 21, 2026+0+00
Thu, Oct 22, 2026+1+00
Total+322+290
  • 1#1 on the list
  • 7Rank that day
  • Not trending

Social mentions

Recent discussions about this repository across the web
  1. Sep 212 posts

    • 目前还没有一个具有多模态功能的jev吗？我看了一些项目 我发现好像没有一个能力很强的，typesafe官方的jev演示案例输入的也是状态信息而非真正的图像。 再观望一下未来会不会有什么多模态方面的进展。

      Read full on X, posted by @Yayoi_no_yume

      @Yayoi_no_yume · x.com

    • 一个用来学 Apple 芯片上视觉推理的教学项目，共享上下文、直接给候选打分、全本地跑，模型是 Qwen3.5-0.8B。 作者自己写了，Breakout 那个例子暴露了 0.8B 四比特模型在这个设置下的极限。

      Read full on X, posted by @aigeeknews

      @aigeeknews · x.com

  2. Sep 191 post

    • 在 Apple Silicon 本机跑通类 Jev 视觉推理，一张图一次回答几十个问题，直接给候选输出打分而不自回归生成。 给 Apple Silicon 写的视觉模型推理练手项目，MLX 跑 Qwen3.5-0.8B，一张图能连答 1 到 64 道题，选项选择、是非、分级打分都行，每题最多 26 个候选。不走自回归写 JSON 的老路，多模态上下文只 prefill…

      Read full on X, posted by @QingQ77

      @QingQ77 · x.com

  3. Sep 181 post

    • 社区版视觉 jev 录制了一个手势识别 demo，本地 16g m4 mbp，200ms 识别手势

      Read full on X, posted by @hr98w

      @hr98w · x.com

  4. Sep 171 post

    • 视觉版 Jev 来了 仿照着社区的思路，把候选结果映射成固定标签，prefill 完直接采样，限制输出 token，采用多模态的 qwen-3.5-0.8B 4bit 量化部署在本地，16g m4 mbp 顺利运行 叠甲：这只是在 infer 层小小的复现一下 jev 的形式，肯定不如各种成熟推理框架效率高，也肯定不是 jev 的正在原理，但是作为 inference 的小入门仍旧是不错的…

      Read full on X, posted by @hr98w

      @hr98w · x.com

You might also like

Repositories close to hr98w/jev-visual
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

Busbar

Control what AI can do before it acts

Blaizzy/mlx-vlm

Python

MLX-VLM is a package for inference and fine-tuning of Vision Language Models (VLMs) on your Mac using MLX.

5.6k840

jjang-ai/vmlx

Python

vMLX - Use MLX models easily - JANGQ (GGUF for MLX) - Not dependant on mlx_vlm

899106

scouzi1966/maclocal-api

Swift

'afm' command cli: macOS server and single prompt mode that exposes Apple's Foundation and MLX Models and other APIs running on your Mac through a single aggregated OpenAI-compatible API endpoint. Supports Apple Vision and single command (non-server) inference with piping as well . Now with Web Browser and local AI API aggregator

34817

incoai/splash

C++

A local inference engine for Apple silicon, built around the model.

1.7k158

apple-aiml-research/ml-fastvlm

Python

This repository contains the official implementation of "FastVLM: Efficient Vision Encoding for Vision Language Models" - CVPR 2025

7.4k561

lmstudio-ai/mlx-engine

Python

LM Studio Apple MLX engine

1.2k139

Meituan-AutoML/MobileVLM

Python

Strong and Open Vision Language Assistant for Mobile Devices

1.4k92

gouwsxander/Apple-Intelligence-API

Swift

A Web API for Apple Intelligence

28033

TheStageAI/edge-lm

Python

Tiny llms optimised for edge deployment

1033