Skip to content
localmodel.run

Text model · llama

Llama 3.2 1B requirements

llama family · 1B params · released Sep 2024 · 59.5M Ollama pulls · LMArena Elo 1110. Minimum to run at Q4_K_M: Apple M1 (8GB).

LicenseLlama 3.2 Community· Conditional↓ 5.5M/mo♥ 1.5Kon HuggingFace
Q4_K_M
0.81 GB
Q8_0
1.32 GB
Total @ Q4 (4k)
~1.8 GB
Context
128 k

Quantization sizes

GGUF quantson disk
QuantizationSize on disk
Q2_K0.4 GB est
Q3_K_M0.5 GB est
Q4_K_M (default)0.81 GB
Q5_K_M0.7 GB est
Q6_K0.8 GB est
Q8_01.32 GB
FP162 GB est

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

Ollama
$ ollama run llama3.2:1b
llama.cpp
$ llama-cli -hf unsloth/Llama-3.2-1B-Instruct-GGUF:Q4_K_M
LM Studio
$ lms get unsloth/Llama-3.2-1B-Instruct-GGUF

Which devices can run Llama 3.2 1B?

FAQ

How much VRAM or RAM does Llama 3.2 1B need?

At Q4_K_M, Llama 3.2 1B needs about 1.8 GB (weights ~0.81 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~2.3 GB.

Can Llama 3.2 1B run on a laptop?

Yes, Llama 3.2 1B fits on a 16 GB machine at Q4_K_M and runs on Apple Silicon or a 12 GB+ GPU comfortably.

Can I use Llama 3.2 1B commercially?

Conditionally. Llama 3.2 Community License: free under 700M MAU.

Released September 25, 2024. Smallest Llama model for edge/on-device use. Ollama tag shows 1.3GB (Q8_0 equivalent default). HF unsloth confirms Q4_K_M at 808MB (~0.81GB) and Q8_0 at 1.32GB.

Sources

Memory figures are estimates. See methodology.