40 articles · newest first
An Android app called RikkaHub Agent turns a phone into a local-LLM-powered agent that can write code, run Linux commands, and compile software on its own — as long as you keep saying yes.
Priya Nair · 18h ago · 8 min
A hobbyist just proved a genuine language model can run entirely on an $8 microcontroller — no cloud, no GPU, no internet connection required.
Ravi Malhotra · 3d ago · 6 min
Phi-4 Mini, Gemma 3, and Llama 3.2 all advertise roughly the same 128K context window. Benchmark data says that number means something very different for each one.
Lena Fischer · 5d ago · 7 min
Four real paths to running a 70B-parameter model on your own hardware or by the hour — priced out with actual 2026 numbers.
Lena Fischer · 6d ago · 8 min
Llama is the open model that kicked off the whole local-AI movement, and running it yourself is genuinely one command. Here's how, and which size fits your GPU.
Lena Fischer · Aug 3 · 10 min
Mistral's models are famously efficient — they punch above their size, so you get a lot on modest hardware. Here's how to run Mistral locally in one command, by size.
Lena Fischer · Aug 3 · 10 min
Qwen3 is arguably the best open model you can run at home right now. Getting it running is genuinely one command. Here's exactly how, and which size fits your GPU.
Lena Fischer · Aug 1 · 10 min
No cloud, no Wi-Fi, no GPU — just a cheap microcontroller quietly telling itself bedtime stories, one token at a time.
Lena Fischer · Jul 31 · 6 min
VRAM is the single number that decides which AI models you can run. Here's the simple rule of thumb, a size-by-size table, and a calculator to check any model against your card.
Ravi Malhotra · Jul 30 · 9 min
Local AI is private, free-per-use, and offline — but the cloud is more capable and needs no hardware. Here's a straight answer on when running your own AI is worth it and when it isn't.
Ravi Malhotra · Jul 30 · 9 min
They're the two most popular ways to run AI models on your own computer, and both are free. The short answer: LM Studio if you want a friendly app, Ollama if you want to build. Here's the full comparison.
Ravi Malhotra · Jul 30 · 9 min
Gemma 3 comes in sizes from tiny to 27B, and each has a very different hardware requirement. Here's exactly what you need to run each one — and which fits the card you already have.
Lena Fischer · Jul 30 · 9 min
Gaming laptops have fast GPUs, but they cap out at 16GB of VRAM. A MacBook Pro's unified memory goes to 128GB. For running AI models on the go, memory wins.
Lena Fischer · Jul 30 · 9 min
You don't need a $3,000 rig to run real local AI. For about a grand you can build a machine that runs 7B models fast and 14B models comfortably. Here's the exact parts list.
Ravi Malhotra · Jul 30 · 9 min
Everyone talks about the GPU price, but a local AI rig has a running cost too. It's smaller than you fear — but if you're in Europe, it's not nothing. Here's the real math.
Ravi Malhotra · Jul 28 · 9 min
Both are excellent open coding models that run on a single card. But they're tuned differently, and the right one depends on how you code. Here's the honest comparison.
Lena Fischer · Jul 28 · 9 min
8GB is the entry floor for local AI, and plenty of capable models run on it if you pick the right size and quantization. Here's what actually works, and the trap to avoid.
Lena Fischer · Jul 28 · 9 min
Both run local models silently and efficiently, but the gap between them for AI isn't the chip — it's the memory bandwidth and how much you can buy. Here's which Mac Mini to get.
Lena Fischer · Jul 28 · 9 min
It's slow, it's old, and it has more VRAM than cards costing twice as much. For a first local-AI GPU, the humble 3060 12GB is still the smart-money entry point.
Ravi Malhotra · Jul 28 · 9 min
Everyone obsesses over VRAM, but system RAM matters too — especially the moment you try to run a model bigger than your GPU. Here's the real number for every kind of local-AI setup.
Lena Fischer · Jul 28 · 9 min
On paper, 12GB for $249 is the best VRAM-per-dollar in new cards. In practice, Intel's software stack is the catch. Here's the honest verdict for local AI.
Lena Fischer · Jul 28 · 9 min
It's the value middle of the Blackwell lineup: 16GB of fast GDDR7, NVFP4 support, and enough speed to run the 14B tier comfortably without paying 5090 money. Is it the right local-AI card for you?
Ravi Malhotra · Jul 28 · 9 min
They all run local models, but they're built for completely different people. One's for easy desktop use, one's for flexibility, one's for serving at scale. Here's how to pick without wasting a weekend.
Lena Fischer · Jul 28 · 9 min
A used 3090 is still the best VRAM-per-dollar card for local AI — but many were mining cards, and the memory runs hot. Here's exactly what to check before you hand over the cash.
Ravi Malhotra · Jul 28 · 9 min
16GB is the awkward middle of local AI: too much for the 8B tier, not enough for 32B. Here's what actually runs well, what to avoid, and how to get the most out of the memory you have.
Lena Fischer · Jul 28 · 9 min
Qwen3-32B at Q4 needs about 20GB of VRAM, which rules out every 16GB card and makes this a simple question: what's the cheapest 24GB GPU that runs it well? The answer is used.
Ravi Malhotra · Jul 28 · 9 min
ExLlamaV2 is open-source and free (MIT), and on a GPU it's roughly twice as fast as llama.cpp. The catch: your model has to fit entirely in VRAM. Here's exactly when to use each one.
Lena Fischer · Jul 27 · 8 min
AMD's Ryzen AI Max+ 395 lets you hand up to 96GB of its unified memory to the GPU — enough to run a 70-billion-parameter model on a mini-PC. Here's exactly how much you can allocate, how to do it, and the catch.
Ravi Malhotra · Jul 27 · 9 min
--n-cpu-moe is the setting that lets a modest graphics card punch far above its VRAM. It offloads the parts of a Mixture-of-Experts model you use least to system RAM — and here's exactly how to tune it for your card.
Lena Fischer · Jul 27 · 9 min
The best open coding model in the world needs a datacenter to run. The best coding model you can run on a single 24GB card is a different name entirely — and it's shockingly close for real work.
Lena Fischer · Jul 24 · 9 min
Kimi K3's open weights land July 27. It's genuinely great and genuinely free. It also needs about 1.4TB of memory to load, which means the download is the easy part. Here's the real hardware math — and the far smaller model that gets you most of the way.
Lena Fischer · Jul 23 · 9 min
Llama 70B needs about 40GB of VRAM to run well — and almost no single consumer card has it. Here's what actually works, visualized, with the measured speeds and real costs for each path.
Ravi Malhotra · Jul 23 · 10 min
Every 'best budget GPU' list is a spec dump. This is a decision: three cards under $500, what each can actually run, and the one number — VRAM — that tells you which to buy without reading a benchmark.
Ravi Malhotra · Jul 22 · 9 min
A palm-sized box that sips power and runs mid-size local models is a genuinely great story — until you ask it to process a long prompt. What the M4 Pro Mac mini is brilliant at, and the one thing it's quietly bad at.
Lena Fischer · Jul 22 · 9 min
Blackwell versus the old flagship, and for once the newer card loses. The 5080's 16GB caps what you can run; the 4090's 24GB doesn't. For local models, capacity beats the spec sheet — with the measured numbers to prove it.
Ravi Malhotra · Jul 22 · 9 min
The RX 7900 XTX gives you 24GB cheaper than a 3090. The catch isn't what you've heard — a llama.cpp scoreboard and an open bug report suggest its Vulkan backend beats ROCm for inference, no painful install required.
Lena Fischer · Jul 22 · 9 min
A new RTX 5060 Ti 16GB looks like the safe budget pick. But 16GB can't load the 32B models a used 3090 runs at 30 tokens/sec — and in local AI, the card that opens the file wins before speed matters.
Ravi Malhotra · Jul 22 · 9 min
Measured at 102.7 tokens/sec — double a 3090 — with the only 32GB of VRAM in consumer land. Then the street price doubles the MSRP and the whole verdict flips. Both answers, with the math.
Lena Fischer · Jul 22 · 9 min
On small-model generation they're within 5% of each other — measured. The real decision lives in what each machine physically can't do, and nobody selling either will tell you.
Lena Fischer · Jul 22 · 10 min
Measured llama.cpp numbers say the 4090 wins every speed test and costs twice as much — while energy per generated token is a dead heat. Here's who each card is actually for.
Ravi Malhotra · Jul 22 · 9 min