Local AI basics
Ollama Not Using GPU? Check Usage and Fix CPU Fallback
Use ollama ps to tell CPU-only inference from partial GPU offloading, then follow focused checks for compatible hardware, drivers, memory, and runtime logs.
Read the guideLocal AI hardware guides
Practical explainers for matching local AI models, context, VRAM, RAM, and runtime settings to the PC you already have.
Local AI basics
Use ollama ps to tell CPU-only inference from partial GPU offloading, then follow focused checks for compatible hardware, drivers, memory, and runtime logs.
Read the guideLocal AI basics
Many PCs can run a small local model. Find your GPU and dedicated VRAM, installed RAM, and OS, then check a specific model and workload with AIPCFit before downloading it.
Read the guideLocal AI basics
There is no single VRAM number that works for every model. Use model size as a first estimate, then account for context, runtime overhead, and the rest of your system.
Read the guideLocal AI basics
Slow local AI does not always mean your computer is too weak. A model may be falling back to the CPU, running out of memory, or using a context setting that is larger than you need.
Read the guideLocal AI basics
Ollama lists Qwen3.8 27B Q4_K_M at 18GB and Q8_0 at 30GB. That makes a 16GB GPU too small for full weight residency; 24GB is a limited-headroom Q4 planning fit, while 32GB gives Q4 more room. These are package-size comparisons and AIPCFit estimates, not measured requirements or speed claims.
Read the guideStart from your hardware
AIPCFit compares your PC profile with local LLM and workload data so memory limits and compatibility gaps are visible before setup.