50 Best Tweets About Local LLMs (2026)

A

Alex Prompter

@alex_prompter · Mar 21, 2026

All-time 11.4K View on X

🚨 BREAKING: Someone just open-sourced a full offline survival computer with AI, Wikipedia, and maps built in. Project N.O.M.A.D. is an open-source offline survival computer. Self-contained. Zero internet required after install. Zero telemetry. Everything runs locally on your … Read more

G

Guri Singh

@heygurisingh · Mar 11, 2026

All-time 9.8K View on X

Holy shit... Microsoft open sourced an inference framework that runs a 100B parameter LLM on a single CPU. It's called BitNet. And it does what was supposed to be impossible. No GPU. No cloud. No $10K hardware setup. Just your laptop running a 100-billion parameter model at … Read more

Watch video

T

Tech with Mak

@techNmak · Mar 15, 2026

All-time 1.5K View on X

Claude Code can run entirely on your local GPU now. Unsloth AI published the complete guide. The setup itself is straightforward - llama.cpp serves Qwen3.5 or GLM-4.7-Flash, one environment variable redirects Claude Code to localhost. But the guide is valuable because of what … Read more

I

Ihtesham Ali

@ihteshamali · Mar 20, 2026

All-time 1.4K View on X

This feels illegal. Someone built a fully local deep research agent that writes its own search queries, hunts down sources, finds the gaps in its own answers, then keeps searching until it's done. It's called Local Deep Researcher. Drop in any Ollama model like DeepSeek, … Read more

Watch video

S

Stan Kirdey

@stan_info · Mar 27, 2026

All-time 878.7 View on X

i’m not a local llm type of guy at all, was just curious and decided to mess around… ended up running a full uncensored qwen3.5-27b (abliterated) on my single 3090 ti with 262k context + tool calling. threw a cloudflare tunnel on it so i can hit the api from anywhere. huge … Read more

O

ollama

@ollama · Mar 17, 2026

All-time 690.2 View on X

Ollama 0.18.1 is here! 🌐 Web search and fetch in OpenClaw Ollama now ships with web search and web fetch plugin for OpenClaw. This allows Ollama's models (local or cloud) to search the web for the latest content and news. This also allows OpenClaw with Ollama to be able to … Read more

J

Jaber

@Akashi203 · Feb 19, 2026

All-time 525.8 View on X

i built a local LLM inference engine that runs a 1B parameter model on a $10 board with 256mb ram. model sits on the sd card, streams one layer at a time through 45mb of ram You can use it as local LLM model backend for PicoClaw no python no cloud no api keys 80kb binary | pure … Read more

0

0xSero

@0xSero · Apr 3, 2026

All-time 478.3 View on X

Here’s what I’d recommend if you’re just getting started in AI, local or otherwise. 1. Work with the compute you have, even the dumbest LLMs can be useful if you treat them as a node in your system. Some basic problems of what could be useful to get you started - tag all your … Read more

A

Akshay 🚀

@akshay_pachaar · Mar 10, 2026

All-time 423.3 View on X

Run Claude Code using local LLMs for FREE. No API costs. No data leaving your machine. Here's how it works: Claude Code lets you swap its backend via a single env variable. Point `ANTHROPIC_BASE_URL` to a local llama.cpp server, and it'll route all requests to whatever model … Read more

A

Ahmad

@TheAhmadOsman · Apr 3, 2026

All-time 377.1 View on X

You don’t pick an Inference Engine You pick a Hardware Strategy and the Engine follows Inference Engines Breakdown (Cheat Sheet at the bottom) > llama.cpp runs anywhere CPU, GPU, Mac, weird edge boxes best when VRAM is tight and RAM is plenty hybrid offload, GGUF, … Read more

A

AI Engineer

@aiDotEngineer · May 21, 2026

All-time 322.4 View on X

TLMs: Tiny LLMs and Agents on Edge Devices with @cormacb https://t.co/u0fHD7j5kZ Function Gemma ships at 270 million parameters and runs nearly 2,000 tokens per second prefill on a Pixel 7. Out of the box, it hits 46% accuracy on a fixed set of app intents. Fine tune on a … Read more

A

Alex Finn

@AlexFinn · Jun 21, 2026

All-time 261.8 View on X

5 months ago I spent $30,000 on 3 Mac Studios, 2 Mac Minis, and a DGX Spark I went all in on local LLMs and encouraged others to do the same I warned prices would explode I was called crazy, a hype beast, dangerous, and that I had no idea what I was talking about Since then: … Read more

W

witcheer

@witcheer · Mar 29, 2026

All-time 258.2 View on X

Google TurboQuant maybe reduced my LLM memory by 6x. running 70B models on my Mac Mini could be real. I have Mac Mini M4. memory was always the issue, 16GB meant 30B models were pushing it, 70B were out of the question. TurboQuant changes that math. 6x memory reduction without … Read more

0

0xSero

@0xSero · Apr 4, 2026

All-time 239.8 View on X

Best harnesses for local models: 1. Droid: - Very good performance, forces the models to behave, you can wire in all your local LLMs very easily w BYOK - Allows you to use your local models as orchestrators/subagents so you can benefit from Cloud as models as well - Practically … Read more

E

Ettore Di Giacinto

@mudler_it · Apr 1, 2026

All-time 237.0 View on X

I've just released APEX (Adaptive Precision for EXpert Models): a novel MoE quantization technique that outperforms @UnslothAI Dynamic 2.0 on accuracy while being 2x smaller for MoE architectures. Benchmarked on Qwen3.5-35B-A3B, but the method applies to any MoE model. Half the … Read more

K

Kshitij Mishra | AI & Tech

@DAIEvolutionHub · Mar 17, 2026

All-time 207.7 View on X

Holy shit 🤯 Microsoft just open-sourced a framework that runs a 100B parameter LLM on a single CPU. No GPU. No cloud. No expensive setup. Just your laptop. It’s called BitNet. And it breaks one of the biggest assumptions in AI. Here’s the trick: Most LLMs use 16-bit or … Read more

Watch video

G

Graham Neubig

@gneubig · Apr 5, 2026

All-time 94.5 View on X

In the spirit of language model freedom, as a weekend project I finally set up my own fully local coding agent: * GMKtek box with AMD Ryzen * Qwen3.5 30B A3B, served with lemonade server+llama.cpp * ngrok to expose the endpoint * OpenHands as the agent, currently through the CLI … Read more

A

Ahmad

@TheAhmadOsman · Apr 4, 2026

All-time 92.1 View on X

Local LLMs, Buy a GPU, and the Case for Cognitive Security AGI? There is no guarantee we reach it anytime soon, if ever. And even if we do, there is certainly no guarantee it will run on your machine. Betting your agency on either assumption is a mistake. What is guaranteed is … Read more

M

Machina

@EXM7777 · Apr 3, 2026

All-time 74.9 View on X

how to set up OpenCode (the privacy-first Claude Code alternative) in 3 minutes: > install: npm i -g opencode-ai > configure your preferred model provider > cd into your project directory > run: opencode it reads your codebase locally, builds context on your machine, and only … Read more

S

Sudo su

@sudoingX · Apr 14, 2026

All-time 62.1 View on X

first local model running on the rog 5090 mobile. gemma 2 2b it q4_k_m through llama.cpp, hermes agent pointing at the local endpoint. sitting on the balcony watching the beach while the model answers prompts. 24gb vram in a portable machine running a full inference stack plus … Read more

W

witcheer

@witcheer · Mar 31, 2026

All-time 58.6 View on X

I run ollama on a Mac Mini for local compression. every message my agent sends passes through a local qwen model to summarise context before it overflows. speed matters because slow compression means slow responses across every cron job. ollama shipped MLX backend for Apple … Read more

M

Muhammad Ayan

@socialwithaayan · May 28, 2026

All-time 57.5 View on X

oh my.. this shouldn't be possible a 1B model that runs inside your browser, beats every model its size, and comes with its own desktop pet. MiniCPM-5 1B just changed the game for on-device AI. here's everything you need to know 🧵

Watch video

V

Vaishnavi

@_vmlops · Apr 19, 2026

All-time 39.7 View on X

SOMEONE RAN A 122B MODEL ON THEIR MACBOOK... IN ONE NIGHT no cloud. no api fees. no data leaving the machine they went through 3 generations: → ollama + proxy: 30 tok/s → llama.cpp + proxy: 41 tok/s → mlx native server: 65 tok/s the breakthrough? they killed the proxy entirely … Read more

S

Sudo su

@sudoingX · Apr 13, 2026

All-time 28.3 View on X

compiling llama.cpp with cuda 12.8 on the rog strix scar 18 5090. this is the mobile 5090, 24gb vram in a laptop. same vram class as the 3090 desktop i run benchmarks on, just portable. i can run these tests from anywhere now. first up carnice-27b vs qwen 3.5 27b dense vs gemma … Read more

K

Kshitij Mishra | AI & Tech

@DAIEvolutionHub · Jul 20, 2026

All-time 27.6 View on X

CHINA JUST DROPPED AN OCR MODEL THAT CHANGES EVERYTHING. A tiny 3B-parameter model can read an entire 100-page PDF in a single pass. No page splitting. No context loss. No cloud APIs. Meet Unlimited-OCR 👇 • Reads full documents with a 32K context window • Scores 93% on OCR … Read more

Watch video

C

CNX Software

@cnxsoft · Mar 30, 2026

All-time 18.9 View on X

Online guide to select the best hardware for local LLM/AI deployments. https://t.co/Q2LAYo5xxx The website relies on Qwen 3.5 models and shows the price, performance, power consumption, and more for a range of hardware from Raspberry Pi 5 16GB to NVIDIA DGX Spark. In one … Read more

M

Maddy 🐝

@Cyb3rMaddy · Jan 16, 2026

All-time 17.1 View on X

Been messing around with local, uncensored LLMs... Ollama runs open-source LLMs locally. It’s handy for security research, private workflows, and anything you don’t want leaving your machine — even works without internet. No cloud calls. No sending prompts somewhere you can’t … Read more

L

Louis Gleeson

@aigleeson · Apr 24, 2026

All-time 14.0 View on X

Claude Code just got a free mode. Not official. A developer built a repo that routes Claude Code calls to free and local AI models. Instead of buying Anthropic credits... You can run it with: > NVIDIA NIM > OpenRouter free models > DeepSeek > LM Studio > llama.cpp Best … Read more

T

Tony Simons

@tonysimons_ · Jul 15, 2026

All-time 11.7 View on X

A 27B multimodal AI model is now running locally on an iPhone. PrismML’s Bonsai 27B packs 27.8B parameters into just 3.9GB and reportedly hits 11 tokens per second on an iPhone 17 Pro. That’s roughly 14× smaller than FP16 while retaining more than 90% of benchmark performance. … Read more

M

Morgan

@morganlinton · Mar 22, 2026

All-time 8.9 View on X

Fun little Sunday morning project, starting work on a little offline LLM I'm calling Wilderness Bot. I go hiking and backpacking out in the wilderness a lot, and want a way to have a little LLM loaded with wilderness medicine info, survival guides, etc. Decided to build in … Read more

S

Shruti Codes

@Shruti_0810 · Jun 17, 2026

All-time 6.9 View on X

A LOCAL LLM RUNNING ON ONE 3090 JUST REPLACED GOOGLE HOME FOR 29 SMART DEVICES WITH ZERO CLOUD. It doesn't just answer questions. Ask, “Do I need a jacket?” and it checks your actual weather sensor before replying. The cloud is becoming optional.

Watch video

Z

ZARA

@HeyZaraKhan · Jun 11, 2026

All-time 6.7 View on X

Nobody expected Apple to open-source this. Apple just open-sourced one of the most interesting AI repositories for developers. It's called Core AI Models. Instead of forcing you to convert and optimize models yourself… Apple provides ready-made export recipes, Python … Read more

M

Morgan

@morganlinton · Mar 27, 2026

All-time 5.8 View on X

Everyone is talking about Kimi and Qwen, but I'm honestly surprised more people aren't talking about models like Trinity from Arcee. I've been doing a deeper dive here and it's pretty interesting, here's a few differences that I'm not sure ppl fully realize. - Qwen and Kimi … Read more

D

Daniel Lemire

@lemire · May 26, 2026

All-time 5.7 View on X

AMD is coming up with its small AI PC (AI Halo). It will compete against NVIDIA DGX Spark. Both look a bit like a mac Mini. Just a tiny box. The AI Halo should cost US$4000, so it is accessible to hobbyists and small IT departement. Set it up with LM Studio with its llama.cpp … Read more

0

0xJiuJitsuJerry

@0xJiuJitsuJerry · Mar 29, 2026

All-time 5.1 View on X

Late night breakthrough: My Mac Mini M4 "Mainframe" is now talking to my @zocomputer through MCP 🧠⚡ What does this mean? Local Gemma3, Nemotron, Qwen3.5 all available as tools No Cloudflare tunnel dependency Zero latency on my AI agent swarm Complete data sovereignty From … Read more

A

Axel Bitblaze 🪓

@Axel_bitblaze69 · Aug 1, 2026

All-time 5.0 View on X

local AI is now useful enough to keep at home imp.. a small machine sitting on your desk can now handle a large part of your everyday AI work for roughly a few dollars in electricity each month, depending on what you run and how often. things like: - drafting
- summarising
- … Read more

U

Ujjwal Chadha

@ujjwalscript · May 29, 2026

All-time 4.5 View on X

Your "Local LLM" and “Free AI” dev setup is a massive WASTE of time and money! The hottest trend on X right now is showing off your "local-first" setup. Developers are buying expensive NVIDIA 5090s, bragging about running Llama-3.2 or Phi-3.5 completely offline, and treating … Read more

S

Shruti Codes

@Shruti_0810 · Jul 3, 2026

All-time 3.6 View on X

The most expensive part of this AI setup isn't the hardware. It's... nothing. A Raspberry Pi 5 with 16GB RAM and a 512GB NVMe is powering: • Local LLMs with Ollama • Claude Code via localhost • Bluetooth analysis • Wi-Fi security testing • Packet capture All from a device … Read more

Watch video

S

Saiyam Pathak

@SaiyamPathak · Apr 1, 2026

All-time 3.4 View on X

Yesterday in the video I took - the qwen32B which was a dense 32B and all active parameters for every token whereas for the MLX version it was A3B - active 3B. this morning I ran some tests again: - Qwen3.5 (NVFP4, MLX): 23.11 tok/s decode - Nemotron (GGUF, llama.cpp): 21.70 … Read more

V

Vaishnavi

@_vmlops · May 20, 2026

All-time 2.1 View on X

CLAUDE CODE FOR FREE - WITHOUT THE ANTHROPIC API BILL Someone built a local proxy that intercepts claude code's api calls & reroutes them to free models point one env variable at localhost and you're done: → nvidia nim (free tier) → openrouter free models → deepseek → ollama / … Read more

D

DEV Community

@ThePracticalDev · May 15, 2026

All-time 1.8 View on X

Arch Linux, the niri scrolling Wayland compositor, llama.cpp with ROCm, and a 27B model running fully offline on 128GB unified memory. This dev shares why local-first AI coding matters and exactly how the stack fits together. { author: @deepu105 } https://t.co/de9zwkVBma

R

Ruslan Khairullin

@Rus_Khairullin · Apr 2, 2026

All-time 1.7 View on X

Vitalik published a detailed post on how he set up a fully local, self-sovereign AI - no cloud, maximum privacy and security. AI agents can already work for hours, use tools and modify their own code. But most (even open-source) ignore security: data leaks, hidden instructions, … Read more

S

smrati tiwari

@smratitiwa86867 · Apr 21, 2026

All-time 1.6 View on X

Holy shit...Someone tried replacing Claude with a local LLM… …and waited 13 minutes for THIS: > “I am a large language model, trained by Google.” That’s it. 13 minutes. One useless sentence. Let’s be honest — we’ve ALL had this thought: “Why am I paying for Claude when I … Read more

D

DBCrypto

@DBCrypt0 · Apr 11, 2026

All-time 1.5 View on X

OMG a new local LLM dropped that you can run on your Mac Mini and it’s just as good as Opus 4.6! 🔥 Local Model: Hi! Before we begin, tell me about yourself and what you want to call me? Me: Let’s call you Max. I’m a content creator, podcast host, and Web3/AI researcher. Local … Read more

J

Jeremy Morgan

@JeremyCMorgan · Jun 17, 2026

All-time 1.5 View on X

If your last local-LLM setup was early 2025, the viable model list moved. This refresh covers Ollama, llama.cpp, and VRAM sizing, with Qwen 2.5 Coder 32B the standout coding pick at 24GB, scoring ahead of GPT-4o on HumanEval per the post. Recalibrate before buying a box. … Read more

A

Aytunc Yildizli

@Aytunc · Apr 29, 2026

All-time 1.2 View on X

⁠A local model, running on my desk, just wrote a game about Erlik, the Turkic underworld god. DeepSeek-V4-Flash (IQ2XXS gguf, llama.cpp after some fight with mlx version) on a Mac Studio. No cloud, no API. Your own model, your own myths, your own game. 2026 is weird.

T

Towards Data Science

@TDataScience · Jun 24, 2026

All-time 0.78 View on X

From installing Ollama to launching OpenCode with a local model, Shuai Guo presents a step-by-step tutorial on building your own local coding agent with Gemma 4. https://t.co/EkLE5CCAOc

J

Julian Goldie SEO

@JulianGoldieSEO · May 6, 2026

All-time 0.45 View on X

This tool feels illegal to be free. Ollama now works with Claude Desktop and Claude Code. That means local models inside your Claude workflow. Your private code can stay on your own machine. Finally, AI without the cloud babysitter. Link in the comments 👇

J

Julian Goldie SEO

@JulianGoldieSEO · Jul 24, 2026

All-time 0.39 View on X

OLLAMA JUST FIXED THE BIGGEST PROBLEM WITH LOCAL AI AGENTS Gemma 4 could use tools before. Now it can actually finish the job. What changed: → Ollama 0.32.1 improves tool-response continuation → Gemma 4 can call a tool, process the result, and continue working → Fewer … Read more

Watch video

T

Towards Data Science

@TDataScience · Jul 26, 2026

All-time 0.17 View on X

To gain a granular understanding of the true costs of running local LLMs, Arsen Apostolov measured the actual GPU electricity for eight models on one RTX 3090 — and produced some counterintuitive findings along the way. https://t.co/qoOaCdek0p