Fits comfortably
Running Llama 3.2 1B Instruct on NVIDIA GeForce RTX 3080 16GB (Mobile)
NVIDIA GeForce RTX 3080 16GB (Mobile) runs Llama 3.2 1B Instruct comfortably at Q4_K_M with 14 GB of headroom for context.
Model size
1B params
Llama 3.2 1B Instruct →Memory available
Recommended quant
Q4_K_M
Highest quality that fits
Quick start with Ollama
1. Install
ollama pull llama3.2:1b2. Run
ollama run llama3.2:1bDefault quant in Ollama is Q4_K_M. To use a different quant, append it: llama3.2:1b-q5_K_M.
Variants and what fits
| Quantization | File size | VRAM required | Fits on NVIDIA GeForce RTX 3080 16GB (Mobile)? |
|---|---|---|---|
| Q4_K_M | 0.8 GB | 2 GB | Yes |
| Q8_0 | 1.3 GB | 2 GB | Yes |
Real benchmarks
Frequently asked
Can NVIDIA GeForce RTX 3080 16GB (Mobile) run Llama 3.2 1B Instruct?
NVIDIA GeForce RTX 3080 16GB (Mobile) runs Llama 3.2 1B Instruct comfortably at Q4_K_M with 14 GB of headroom for context.
What quantization should I use?
Q4_K_M is the highest-quality variant of Llama 3.2 1B Instruct that fits in 16 GB VRAM. Lower-bit quants will be smaller but lose some quality.
How fast will it be?
Measured at 189.5 tok/s on this combination in our testing.
See also: Llama 3.2 1B Instruct, NVIDIA GeForce RTX 3080 16GB (Mobile), all benchmarks.
Reviewed by RunLocalAI Editorial. See our editorial policy.