Free tools Windows power users keep installed
One-click scans. No signup required.
To run a coding model locally, install an inference runtime, download model weights it supports, and load a model that fits your computer’s memory. Choose LM Studio for a graphical workflow, Ollama for a simple command line and local API, or llama.cpp for more control over model files and compute backends. The steps below cover setup, hardware fit, and connecting compatible coding software.
Choose a local runtime
A runtime loads model weights and performs inference on your computer. It is separate from the model itself: installing the runtime does not automatically provide the weights you need. Compare the setup style and degree of control before choosing.
| Runtime | Setup style | Model files and control | Local API |
|---|---|---|---|
| LM Studio | Graphical app: find and download a model in Discover, then load it in Chat. | Provides an in-app model discovery workflow; supports model weights in formats such as GGUF and safetensors. | Documents local REST and OpenAI-compatible APIs. |
| Ollama | Terminal commands for downloading and running models. | Manage models using Ollama commands; model catalog and sizes may change. | Documents a local REST API for generating or chatting. |
| llama.cpp | Command-line runtime, installed through a package manager, Docker, prebuilt release, or source build. | Requires GGUF files; supports quantization and CPU/GPU hybrid inference. | Provides llama-server, an OpenAI-compatible server. |
No universal speed or coding-quality ranking is established by these product documents. The best starting point depends on your operating system, memory, preferred workflow, coding tasks, and the model’s license.
Check hardware and storage before downloading
There is no single minimum that applies to every local model. Memory needs vary with model size, quantization, context length, runtime, and how much work can be offloaded to the GPU. RAM and dedicated GPU memory are distinct resources; an SSD can hold model files but cannot replace memory needed while the model runs.
#1 Best Overall
- EVOLUTION AMD RYZEN AI MAX+ 395 MINI PC - GMKtec EVO-X2 is the next evolution in AI mini PC Ryzen Strix Halo series. Thanks to AMD Simultaneous Multithreading (SMT) the core-count is effectively doubled, to 32 threads. Ryzen AI Max+ 395 has 64 MB of L3 cache and can boost up to 5.1 GHz, depending on the workload. The Ryzen AI Max+ 395 is currently rated as the "most powerful x86 APU" on the market for AI computing.
- AI NPU with XDNA 2 ARCHITECTURE - Powered by 16 “Zen 5” CPU cores, 50+ peak AI TOPS XDNA 2 NPU and a truly massive integrated GPU driven by 40 AMD RDNA 3.5 CUs, the Ryzen AI MAX+ 395 is a transformative upgrade and delivers a significant performance boost over the competition. The Ryzen AI Max+ 395 excels in consumer AI workloads like the llama.cpp-powered application: LM Studio. Shaping up to be the must-have app for client LLM workloads, LM Studio allows users to locally run the latest language model without any technical knowledge required and unleash their creativity and productivity.
- AMD RADEON 8090S iGPU GAMING PC - The AMD Radeon RX 8060S offers all 40 CUs with up to 2.9 GHz graphics clock and uses the new RDNA 3.5 architecture. The powerful iGPU is positioned between an RTX 4060 and 4070 laptop GPU and therefore enables gaming in FHD at maximum details in most demanding games. The 8060S can also utilize the full 128GB pool, which is perfect for running LLMs such as Deepseek 70B Q8, which runs comfortably on this machine.
- EIGHT CHANNEL LPDDR5X - LPDDR5X is a new ground breaking memory small form factor installed on-board. With blazing speeds up to to 8000MT/s, it runs 1.5x faster than the DDR5 SODIMMs; 90% better performance over DDR5 SODIMMs in video conferencing and photo editing; 30% better performance in productivity apps; 12% better performance in digital content workloads.
- QUAD SCREEN 8K DISPLAY SUPPORT - EVO-X2 AI Mini PC support 4-screen 4K/8K output via HDMI 2.1 (8K@60Hz), DisplayPort 1.4 (4K@60Hz), and dual USB 4 40Gbps Transfer speed (supporting PD3.0/DP1.4/DATA). Ideal for gaming, video editing, and multitasking, it provides expansive and crisp multi-display support.
LM Studio’s stated requirements
LM Studio’s system requirements page, accessed in 2026, recommends at least 16GB RAM for Apple Silicon Macs; it says an 8GB Mac may still work with smaller models and modest context sizes. For Windows, it recommends 16GB RAM and at least 4GB of dedicated GPU VRAM, and x64 requires AVX2. The page lists Windows x64 and ARM, Linux x64 and ARM64, and macOS 14 or newer on Apple Silicon M1, M2, M3, or M4. These are LM Studio’s requirements and recommendations, not universal requirements for other runtimes. See LM Studio System Requirements.
Ollama’s memory rules of thumb
Ollama’s quickstart suggests at least 8GB available RAM for 7B models, 16GB for 13B models, and 32GB for 33B models. Treat these as Ollama guidance, not a guarantee for every quantization, context length, or machine. The same page gives illustrative download sizes: Llama 3.2 1B at 1.3GB, Llama 3.2 3B at 2.0GB, Llama 3.1 8B at 4.7GB, and Llama 3.1 70B at 40GB. Those are model-file sizes, not total runtime memory requirements. Check the current catalog and machine fit on the Ollama Quickstart.
Rank #2
- Built for Local AI Development: AMD Ryzen AI Halo is designed for local AI development and inference, featuring 128GB unified memory and support for up to 200B parameter models to build and run intensive AI workloads locally.
- 128GB Unified Memory: Features 128GB LPDDR5x unified memory at 8000 MT/s with 256 GB/s memory bandwidth, providing a shared memory pool across the CPU, GPU, and NPU to support larger AI models.
- AMD Ryzen AI Max+ 395 Processor: Features 16 cores, 32 threads, and Zen 5 architecture, paired with AMD Radeon 8060S integrated graphics featuring 40 RDNA 3.5 compute units and an AMD XDNA 2 NPU with up to 50 TOPS.
- Linux AI Developer Platform: Purpose-built for Linux-based AI development with full AMD ROCm software support and preloaded tools, models, and workflows optimized for local AI development.
- Compact, Connected Design: Includes a 2TB M.2 SSD, 10GbE LAN, Wi-Fi 7, Bluetooth 5.4, USB-C connectivity, and HDMI 2.1b.
Quantization and disk space
Quantization stores model weights in a reduced-precision representation, often lowering memory use while potentially affecting output quality. llama.cpp documents levels from 1.5-bit through 8-bit and CPU/GPU hybrid inference, but there is no single ideal quantization or model size for coding across all hardware. Start with a model that fits, then evaluate it on your own coding tasks.
Plan disk space separately from RAM and VRAM. Keeping several models can consume substantial storage; an external SSD may help if internal storage is limited, but no universal capacity or speed requirement is established here. It does not substitute for system or GPU memory while a model is running.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Rank #3
- EVOLUTION AMD RYZEN AI MAX+ 395 MINI PC - GMKtec EVO-X2 is the next evolution in AI mini PC Ryzen Strix Halo series. Thanks to AMD Simultaneous Multithreading (SMT) the core-count is effectively doubled, to 32 threads. Ryzen AI Max+ 395 has 64 MB of L3 cache and can boost up to 5.1 GHz, depending on the workload. The Ryzen AI Max+ 395 is currently rated as the "most powerful x86 APU" on the market for AI computing.
- AI NPU with XDNA 2 ARCHITECTURE - Powered by 16 “Zen 5” CPU cores, 50+ peak AI TOPS XDNA 2 NPU and a truly massive integrated GPU driven by 40 AMD RDNA 3.5 CUs, the Ryzen AI MAX+ 395 is a transformative upgrade and delivers a significant performance boost over the competition. The Ryzen AI Max+ 395 excels in consumer AI workloads like the llama.cpp-powered application: LM Studio. Shaping up to be the must-have app for client LLM workloads, LM Studio allows users to locally run the latest language model without any technical knowledge required and unleash their creativity and productivity.
- AMD RADEON 8090S iGPU GAMING PC - The AMD Radeon RX 8060S offers all 40 CUs with up to 2.9 GHz graphics clock and uses the new RDNA 3.5 architecture. The powerful iGPU is positioned between an RTX 4060 and 4070 laptop GPU and therefore enables gaming in FHD at maximum details in most demanding games. The 8060S can also utilize the full 64GB pool, which is perfect for running LLMs such as Deepseek 32B, which runs comfortably on this machine.
- EIGHT CHANNEL LPDDR5X - LPDDR5X is a new ground breaking memory small form factor installed on-board. With blazing speeds up to to 8000MT/s, it runs 1.5x faster than the DDR5 SODIMMs; 90% better performance over DDR5 SODIMMs in video conferencing and photo editing; 30% better performance in productivity apps; 4% better performance in digital content workloads.
- QUAD SCREEN 8K DISPLAY SUPPORT - EVO-X2 AI Mini PC support 4-screen 4K/8K output via HDMI 2.1 (8K@60Hz), DisplayPort 1.4 (4K@60Hz), and dual USB 4 40Gbps Transfer speed (supporting PD3.0/DP1.4/DATA). Ideal for gaming, video editing, and multitasking, it provides expansive and crisp multi-display support.
Install and run a model
LM Studio: graphical setup
- Install LM Studio for a supported operating system, checking its current requirements first.
- Open the Discover tab, find a model, and download its weights. LM Studio names Qwen, Mistral, Gemma, and gpt-oss as examples; availability and model fit can change.
- Open the Chat tab and load the downloaded model into memory. Loading allocates memory for weights and other parameters.
- Enter a coding prompt in chat. To use another application, configure it for LM Studio’s local REST or OpenAI-compatible API and confirm that the client supports the endpoint and model features you need.
LM Studio says offline use is possible once model files have been obtained. See Get started with LM Studio and LM Studio Docs.
Ollama: terminal setup
- Install Ollama using its current instructions for your operating system.
- Download a model with
ollama pull MODEL_NAME, or run it directly withollama run MODEL_NAME. For example, the quickstart documentsollama run llama3.2; treat that as a command example, not a permanent recommendation or catalog guarantee. - Use the interactive prompt to ask coding questions. Use
ollama listto see downloaded models andollama psto inspect running models. - For a coding application, configure it to use Ollama’s local REST API and verify that the client supports the API and required model features.
Ollama’s quickstart documents a local REST API for generation and chat. Model names and sizes can change, so check the current catalog before choosing.
Rank #4
- 【Leading AI Mini Workstation】MINISFORUM AI MS-S1 Max Workstation comes with AMD Ryzen AI Max+ 395 processor, which uses AMD's latest generation Zen 5 architecture. It has 16 Cores and 32 Threads, the boost clock is up to 5.1GHz. The overall processor performance is up to 126 TOPS, and the NPU performance reaches up to 50 TOPS. AMD Ryzen AI enables improved productivity, advanced collaboration, and improved efficiency.
- 【AMD Radeon 8060S Graphics 】The MS-S1 Max Mini PC equipped with AMD Radeon 8060S Graphics which built on the new generation of RDNA 3.5 architecture AMD graphics, it brings ultra-high frame rate experiences and advanced content creation features anywhere and delivers staggering performance. It can handle all your computing and multimedia tasks efficiently.
- 【Five 8K Video Output】This MS-S1 Max Workstation comes with five video outputs, 1x HDMI (8K@60Hz), 2x USB4(40Gbps,Alt DP2.0,PD out 15W) and 2x USB4 V2(80Gbps,Alt DP2.0,PD out 15W) Outputs, which support multiple monitors display at the same time and provide a larger and wider filed of view and improve your work efficiency. It is used in fields that require high-performance computing and graphics processing, including digital signage and securities trading, as well as work that uses CAD, such as engineering design, scientific calculations, animation production, and post-production for movies and television
- 【 Fast and Stable Wire & Wireless Speed】It comes with Two 10G Lan Ports for wired connection and and Wi-Fi 7 / BT5.4 for wireless connection, which increased the network speed greatly and expand its functions and improved performance of computer to a large extent and allows you to use more networks such as software routers (OpenWRT / DD-WRT / Tomato etc.), firewalls, NAT, network isolation etc.
- 【Large Storage & Flexible Expandability】This Workstation equipped with 64GB LPDDR5-8000MHz + 2TB M.2 2280 PCIe4.0 SSD. There is another PCIe4.0 SSD slot available for up to 8TB, these SSD slots are compatible with RAID0 and RAID1, you can store movies, videos, photos, important files easily. What’s more, it also comes with 1x standard PCIex16 slot(PCIe4.0x4) inside.
llama.cpp: direct model-file control
- Install llama.cpp using a documented route: package manager, Docker, prebuilt release, or building from source.
- Obtain a compatible GGUF model file. The README also shows a model-download route using
-hf. - Run a local file with
llama-cli -m my_model.gguf, substituting the actual GGUF file path. - To serve a model to another application, start
llama-serverand configure the client for its OpenAI-compatible API. Check the client’s endpoint and feature requirements.
For supported options and backend details, consult the llama.cpp README.
Connect a coding tool to the local model
All three runtimes document local API routes, and LM Studio and llama.cpp specifically describe OpenAI-compatible APIs. A client may still need configuration for the local endpoint, model interface, and any tool-calling or code-editing features it expects. The existence of an API does not mean every editor extension or coding agent will work without setup; check the chosen client’s compatibility documentation.
Recommended Free Tools
Check model licensing and evaluate it on your work
Model weights are distributed under different licenses. Read the license for the specific model you download and confirm that its terms suit your intended use; the label “open” alone does not establish uniform rights. Then test the model on representative tasks—such as explaining a function, drafting a small change, or diagnosing an error—before relying on it for your workflow. Results depend on the model and task, and the documentation cited here does not establish a coding-quality or performance winner.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
