Run AI on Your Own Hardware
No cloud. No subscription. No data leaving your machine. This section covers everything you need to run, configure, and fine-tune AI models locally โ from a $400 mini PC to a $2000 GPU workstation.
100% private
Your prompts and data never leave your machine. No API logs, no training on your conversations.
No subscription
Pay once for hardware and run unlimited queries for free. One GPU pays for itself in weeks.
Works offline
Run a full LLM on a plane, in a bunker, or on an air-gapped server. No internet required.
Where do you want to start?
Step-by-step guides
All local AI guidesMiniCPM5-2B: Install and Run OpenBMB's 2B Model That Outscores 4B Rivals
MiniCPM5-2B is OpenBMB's 2.52 billion parameter model, released under Apache 2.0, and it averages 53.9 on OpenBMB's own benchmark set, ahead of Qwen3.5-4B's 51.1 despite having roughly half the parameters. Here is how to install it through vLLM, SGLang or Transformers, what the GGUF builds actually weigh, and what the benchmark table does and doesn't prove.
BERT Base Uncased: Specs, How to Run It, and Why It's Still Trending in 2026
google-bert/bert-base-uncased is a 110 million parameter masked-language model from 2018 that just re-entered Hugging Face's overall trending top 20. It has 48.8 million downloads, apache-2.0 licensing, and runs comfortably on a CPU. Here is how to run it, what it's actually still useful for, and why an eight-year-old model outranks this week's releases.
How to Run tencent/AuK Locally: Zero-Shot TTS, Voice Cloning and Speech Editing
tencent/AuK is a 1.5 billion parameter open source speech model that handles zero-shot voice cloning, instruction-guided TTS, pitch and speed editing, and audio source separation through one natural-language interface. It also depends on a separate 3 billion parameter Qwen2.5-Omni encoder at runtime, which changes the real hardware math. Here is what the specs actually are, how to install it, and how to run your first zero-shot clone.
GGUF vs MLX vs NVFP4: Local AI Quantization Formats Explained
Three names keep showing up when you go to download a local model: GGUF, MLX, and the newer NVFP4. They are not interchangeable, and picking the wrong one wastes memory or leaves your hardware idle. Here is what each format is, which one your machine actually wants, and how to choose in ten seconds.
Stable Diffusion 3.5 Medium: What's Different From SDXL and Flux
Stable Diffusion 3.5 Medium is Stability AI's 2B-parameter MMDiT-X text-to-image model. Here is how it differs from the SDXL and Flux coverage already on this site, how to install it with Diffusers, and the licensing catch that kicks in above $1M in revenue.
How to Set Up Local AI on 8GB of VRAM
An 8GB GPU runs the 7B to 8B model class comfortably at Q4 quantization, and Phi-4 14B just barely fits. Here is exactly what to install, which quantization format to pick, and real measured tokens-per-second so you know what to expect before you download anything.
How to Set Up Local AI on 12GB of VRAM
12GB of VRAM is the mixture-of-experts sweet spot: OpenAI's gpt-oss-20b needs about 12GB at Q4 and, because only 3.6B of its 20B parameters activate per token, generates at nearly 7B speed. Here is how to install it, which quantization it actually ships in, and cited real-world tokens-per-second.
Best Local LLM for 16 GB VRAM: Setup, Quantization and Real Speed
16GB of VRAM unlocks Gemma 3 27B, which needs about 15GB at Q4 and just fits, plus OpenAI's gpt-oss-20b with real headroom to spare. Here is how to install both, which quantization format to use, and measured tokens-per-second on the RTX 4060 Ti 16GB and RTX 5070 Ti.
Models to run locally
21 modelsEvery model is free to download and run. Requirements listed are for the Q4 quantized version โ the best balance of quality and size.
Llama 3.1 8B
Meta ยท Llama 3.1 Community
Meta's 8B workhorse with a massive 128K context window. Punches well above its weight class for everyday coding and chat tasks, runs comfortably on any consumer GPU with 6 GB VRAM, and delivers near-70B quality on many benchmarks.
Min VRAM
5 GB
Context
128K
Run with Ollama
ollama run llama3.1:8bLlama 3.3 70B
Meta ยท Llama 3.3 Community
Meta's flagship open-weight model and the go-to choice for users who want GPT-4-class performance without sending data to the cloud. Consistently tops open-source leaderboards. Needs ~48 GB VRAM at Q4_K_M โ a dual-GPU rig or Apple Silicon with 48 GB unified memory.
Min VRAM
48 GB
Context
128K
Run with Ollama
ollama run llama3.3:70bLlama 3.2 3B
Meta ยท Llama 3.2 Community
The smallest production-quality Llama model, designed to run on laptops, Raspberry Pis, and other edge devices with minimal VRAM. Response latency is sub-second on modern CPUs, making it ideal for real-time assistants and local automation scripts.
Min VRAM
2 GB
Context
128K
Run with Ollama
ollama run llama3.2:3bQwen3 8B
Alibaba ยท Apache 2.0
One of the most-downloaded mid-size AI models on HuggingFace in 2026, with tens of millions of downloads. Qwen3 8B outperforms the previous Qwen2.5 14B on most benchmarks, supports hybrid thinking mode and tool calling, and runs on any GPU with 6 GB VRAM. Apache 2.0 license โ fully commercial.
Min VRAM
5 GB
Context
128K
Run with Ollama
ollama run qwen3:8bMistral 7B v0.3
Mistral AI ยท Apache 2.0
The model that proved 7B parameters can match much larger models on reasoning tasks when trained carefully. Apache 2.0 licensed, so it is fully free to use commercially with no restrictions.
Min VRAM
4 GB
Context
32K
Run with Ollama
ollama run mistral:7bMixtral 8x7B
Mistral AI ยท Apache 2.0
A sparse mixture-of-experts model that activates only 12B parameters per token while carrying 47B total, delivering 70B-class results at a fraction of the compute cost. One of the best open-weight models available under a fully permissive license.
Min VRAM
26 GB
Context
32K
Run with Ollama
ollama run mixtral:8x7bBest local LLM for your hardware
Already know your VRAM budget, GPU or Mac? These pages list every model from the verified catalog that fits, with headroom, run commands and spec-sheet links.
Hardware picks
Full buyer's guideThese are affiliate links. We earn a small commission if you buy through them โ at no extra cost to you. Disclosure
NVIDIA RTX 5090 32GB
32 GB GDDR7, 512-bit bus, 575W TDP, 21,760 CUDA cores
Most VRAM on any consumer card โ the current flagship for 34Bโ70B local inference
Runs these models
NVIDIA RTX 5080 16GB
16 GB GDDR7, 256-bit bus, 360W TDP, 10,752 CUDA cores
Current-gen 16 GB card with GDDR7 speed for fast 7Bโ34B inference
Runs these models
NVIDIA RTX 5070 Ti 16GB
16 GB GDDR7, 256-bit bus, 300W TDP, 8,960 CUDA cores
Best current-gen value โ 16 GB GDDR7 for 7Bโ34B at a mid-tier price
Runs these models
NVIDIA RTX 4060 Ti 16GB
16 GB GDDR6, 128-bit bus, 165W TDP, 4,352 CUDA cores
Best VRAM-per-dollar entry card for running 13B models locally
Runs these models
NVIDIA RTX 4070 Super 12GB
12 GB GDDR6X, 192-bit bus, 220W TDP, 7,168 CUDA cores
Sweet-spot performance for fast 7B and 13B token generation
Runs these models
NVIDIA RTX 4080 Super 16GB
16 GB GDDR6X, 256-bit bus, 320W TDP, 10,240 CUDA cores
High-throughput inference for 7Bโ34B models at serious speed
Runs these models
Weekly local AI drops
New models, what runs on your hardware, and the guides to set them up. One email a week, unsubscribe any time.
Affiliate disclosure
Hardware links on this page go to Amazon via our affiliate program. We earn a small commission when you buy through them. This does not change the price you pay or influence our recommendations โ every pick is chosen based on VRAM, value, and real performance for local AI workloads.