← CanIRunLocal · All hardware · All models

Can I run Gemma 3 1B locally?

Google · 1B parameters · chat.

Ollama weight files: Q4_K_M 0.81 GB (gemma3:1b); Q8_0 1.1 GB (gemma3:1b-it-q8_0).

Yes on 62 machines in this list. The smallest is M1 8GB: Q8_0, about 1.8 GB, roughly 49 tokens per second, via Ollama. ollama run gemma3:1b-it-q8_0

• DGX Spark 128GB: Q8_0, ~1.8 GB, fits in fast memory, ~186 tok/s — ollama run gemma3:1b-it-q8_0
• Mac Studio M5 Max 128GB: Q8_0, ~1.8 GB, fits in fast memory, ~447 tok/s — ollama run gemma3:1b-it-q8_0
• Mac Studio M5 Ultra 256GB: Q8_0, ~1.8 GB, fits in fast memory, ~873 tok/s — ollama run gemma3:1b-it-q8_0

Setup on DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte)

• Ollama: ollama run gemma3:1b-it-q8_0
• LM Studio: lms get unsloth/gemma-3-1b-it-GGUF@Q8_0
• llama.cpp: llama-server -hf unsloth/gemma-3-1b-it-GGUF:Q8_0 -c 8192
• vLLM: vllm serve google/gemma-3-1b-it --max-model-len 8192 --gpu-memory-utilization 0.9

Hardware that fits

Estimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-16.

A project by OnDevice Labs · llms.txt