← Blog

llama.cpp

AI

/dictionary/llama-cpp

Definition

A C++ implementation of LLM inference designed to run quantized models on consumer hardware (CPU, CUDA, Metal, Vulkan). The de-facto local inference engine that Ollama and LM Studio both wrap. Supports GGUF format, has a built-in HTTP server, and is the reference for what local LLMs can actually do.

Related terms

Posts that use this term

  • Troubleshooting local LLMs (and how to keep up after this series)

    The full catalog of local-LLM failures: OOM, slow tok/s, garbage output, instruction drift, bad RAG hits, tool-call hallucination. Plus where to follow the field once you're on your own.

  • Fine-tuning a model locally

    When fine-tuning is actually the right call (it usually isn't) and how to pull off a LoRA run on a 16GB Mac, with a worked Llama 3.2 3B example.

  • Wiring a local LLM into the tools you already use

    How to point VS Code (Continue, Cline), web chat UIs (Open WebUI, LibreChat, Page Assist), and your own code at a local model using the OpenAI-compatible API. Swap cloud for local without rewriting anything.

  • Your first local LLM, start to finish

    Install Ollama, pull Llama 3.2 3B, chat with it, hit its API, and fix the five things that break on a first install. You finish with a working local LLM.

  • System requirements by OS for local LLMs

    What macOS, Linux, and Windows each need before you run a local LLM in 2026. Mac is the smoothest, Linux gives you the most knobs, and native Windows finally just works.

  • Picking a local model by task

    The 2026 open leaders, sorted by what you actually want to do: coding, chat, the small-model crowd, structured output, vision, embeddings, and audio.

  • Streaming, throughput, and the KV cache

    Why TTFT and tok/s are different numbers, why streaming feels faster than it is, and the KV cache that makes the 1000th token cost about the same as the first.

  • Quantization, distillation, pruning: how a 140GB model fits on your laptop

    Three ways to shrink an LLM, and why one of them does almost all the work. What Q4_K_M actually means and what each shortcut costs you.

  • The local-LLM vocabulary

    Parameters, B, dense vs MoE, base vs instruct, tokens, context windows, chat templates, GGUF, and quant suffixes. Read it once and any HuggingFace model card stops being scary.

  • The runtimes: llama.cpp, Ollama, LM Studio

    llama.cpp is the engine. Ollama and LM Studio wrap it. What each one does, when to reach for which, and why the OpenAI-compatible APIs are mostly but not entirely interchangeable.

  • Why Apple Silicon punches above its weight on local LLMs

    Unified memory lets the GPU see all of RAM. Here's why that beats a discrete-GPU PC past 32B parameters, what fits in 16/32/64/128/192GB, and where Apple Silicon still loses.

  • What it takes to run a model on your own machine

    Why VRAM is the one number that decides whether a local LLM runs, what quantization really does to a model file, and the hardware ladder from an 8GB laptop to a 192GB workstation.

  • Where AI actually runs: cloud, local, edge

    When you use AI, a model file is sitting on a real machine. There are only three places it can be, and which one decides almost everything else.

  • Install LM Studio

    Install LM Studio on macOS, Linux, and Windows, then flip on the local OpenAI-compatible server so any client library can talk to a model on your own machine.

  • Install llama.cpp

    Build llama.cpp from source with Metal or CUDA, then run a GGUF model with llama-cli. The closest thing to bare-metal local inference.