New in Local AI
6 models, 16 tools and 1 device released in the last 6 weeks. Updated Mon, Thu & Sat; last update Nov 1.
Week of October 26, 2026
1 model
- Ling 3.1 Flash 1019Model
572B params · ~332 GB at Q4 · mit
Week of October 19, 2026
2 tools
EAGLE-4ToolOfficial Inference Code of EAGLE-4
ming-voiceToolLow-latency local speech-to-speech voice assistant with Ming-3.0-audio-tiny on AMD and Nvidia GPUs. Easy setup on Windows and Linux
Week of October 12, 2026
3 models, 1 tool, 1 device
- NVIDIA RTX Spark N1X (128 GB)Hardware
128 GB memory · 273 GB/s
763B params · ~443 GB at Q4 · text + vision · mit
65.8B params · ~38 GB at Q4 · text + vision · other
- MiniMax-M3.1-FlashModel
221B params · ~128 GB at Q4 · text + vision + video · other
hobToolAn easier way to use AI on your own computer. A free, open-source alternative to Claude and ChatGPT that runs open-weight models like Gemma and gpt-oss locally…
Week of October 5, 2026
5 tools
DigUpToolSearch your computer by what's in your files: photos, PDFs, documents, recordings, videos and code. On-device on Mac, Windows and Linux, built on Google DeepMi…

Let an AI model use a real iPhone with a $7 ESP32: AirPlay eyes, USB HID hands, any OpenAI-compatible vision model. No Mac, no developer account, no jailbreak.
project-mayaToolProject Maya - run GLM-5.3-Flash (321B MoE), DeepSeek-V4.1-Flash-1017 (748B MoE) and MiniMax-M3.1-Flash (428B MoE) privately on your own GPU(s), from one card…
PolyStrataToolRun large AI models on your own PC: Qwen3.8-Flash-Next on Strata's engine, and GLM 5.3 Flash, DeepSeek V4 Flash, LFM2.5, Qwen3.6, Ornith 1.5, Gemma 4, Tiel Cod…
XolotlToolXolotl (formerly Strata-DSV4.1): run DeepSeek-V4.1-Flash, all 552B, as a local coding agent on one gaming PC (RTX 4090 + 90 GB RAM) at up to ~27 tok/s, 512K co…
Week of September 28, 2026
5 tools
folkbenchToolAI API 中转站评测排行与切换工具:实测可用率、延迟与价格 · Open-source AI API relay benchmark, rankings & switcher for Claude Code / Codex
dossierToolA self-hosted job search app that writes tailored resumes and cover letters from your profile.

Open decision models from 0.8B to 27B: a calibrated probability for every option in one pass, and a short thought only when unsure. Engine for vLLM, llama.cpp…
backburnerToolYour iPhone helps your Mac run a 27B model: faster prompt reading and more context over a USB-C cable

Your AI agent browses the web without captchas: self-hosted Python MCP server on undetected anti-detect stealth Firefox, for browser automation, scraping, comp…
Week of September 21, 2026
2 models, 3 tools
- MiMo-V2.6-FlashModel
311B params · ~180 GB at Q4 · text + vision + audio + video · mit
- MiMo-V2.6-ProModel
159B params · ~92 GB at Q4 · mit

在 dsh 里装上这个插件即可,无需登录、注册或填 API Key,就能使用包括 DeepSeek V4.1 Flash、Kimi K3 在内的前沿模型——完全免费,不限量。 All you do is install this plugin in dsh: no login, no sign-up, no API…
magpieToolEvery agent's model. One place. Codex on DeepSeek, Claude Code on Kimi, from the menu bar.
lemo-opuscarToolA toolbox of film styles made in code for Claude Code: 66 ready to use, and your own. Each style is a style prompt plus a demo film made entirely in code by Cl…
Dates are public release dates (for tools, when the project first appeared on GitHub). Check what fits your machine on What can I run?, or browse every model, tool and device.