8 Go
Best local LLM for a laptop (8 GB)
Small models that run on a laptop with no dedicated GPU, in Q4, within 8 GB of RAM. Not benchmark champions, but good enough to summarise, rewrite, classify and answer offline.
| # | Model | Arena Elo | RAM (Q4) | Hosting | Licence |
|---|---|---|---|---|---|
| 1 | 🇺🇸Gemma 3n E4B Google | 1317 | 7 Go | Self-hostable | gemma |
| 2 | 🇺🇸Granite 4.1 8B IBM | 1306 | 7 Go | Self-hostable | apache-2.0 |
| 3 | 🇺🇸Gemma 3 4B Google | 1303 | 5 Go | Self-hostable | gemma |
| 4 | 🇫🇷Ministral 8B Mistral AI | 1237 | 7 Go | Self-hostable | mistral-research |
| 5 | 🇺🇸Llama 3.1 8B Meta | 1211 | 7 Go | Self-hostable | llama-3.1-community |
| 6 | 🇺🇸Llama 3.2 3B Meta | 1166 | 4 Go | Self-hostable | llama-3.2-community |
| 7 | 🇺🇸Llama 3.2 1B Meta | 1111 | 3 Go | Self-hostable | llama-3.2-community |
| 8 | 🇫🇷Mistral 7B Mistral AI | 1110 | 7 Go | Self-hostable | apache-2.0 |
| 9 | 🇨🇳Qwen3 0.6B Alibaba | — | 2 Go | Self-hostable | apache-2.0 |
| 10 | 🇨🇳Qwen3 8B Alibaba | — | 7 Go | Self-hostable | apache-2.0 |
Data verified on 4 September 2026Open in the comparator
Method
Deterministic ranking on comparator data: LMArena Elo (style control) first, HuggingFace downloads second for models without an Elo. Prices and licences are verified at the official source and dated. No model is ranked by an LLM.