Bobbin
← Models

Llama 3.2 3B Instruct

Small, fast, runs on almost anything

metaLlama 3.2 Community LicenseFreeReleased September 2024 · 2 years old
Install with Bobbin

Don't have the app? Download it here.

Meta's compact instruct model. The safe first download: it loads in seconds, fits comfortably on 6GB cards, and is genuinely useful for summarising, drafting and quick questions.

Pick something larger if you want deeper reasoning or code.

Available versions

QuantizationSizeVRAMRAMContext
Q4_K_M2.0 GB3.3 GB5.9 GB8kInstall
Q8_03.4 GB4.7 GB7.8 GB8kInstall

Memory needed grows with the length of a conversation, so these figures are for the context shown. Anything marked est. is worked out from the file size rather than measured on real hardware — the app says so there too. The desktop app checks all of it against your actual machine before installing.