Llama 3.2 3B

At the default Q4_K_M, 139 of 140 phones have enough estimated usable memory; 139 also reach 3+ tokens/s. On Mac, 54 of 54 exact configurations pass using that quant or a smaller tracked fallback.

Params
3.2B
Family
llama
Released
2024-09
Tags
chat

Downloads & memory by quant

The download is only the weights — add KV cache and runtime to get what your phone actually needs. “Memory-fit” only means it can load; “usable” also requires an estimated 3+ tokens/s. Each row uses that row's quant. ★ = recommended.

QuantDownloadMemory neededPhone reality
IQ4_XS1.8 GB~2.7 GB139 memory-fit · 139 usable
Q4_01.9 GB~2.8 GB139 memory-fit · 139 usable
Q4_K_S1.9 GB~2.8 GB139 memory-fit · 139 usable
Q4_K_M2 GB~2.9 GB139 memory-fit · 139 usable
Q5_K_M2.3 GB~3.2 GB139 memory-fit · 139 usable
Q6_K2.6 GB~3.6 GB139 memory-fit · 139 usable
Q8_03.4 GB~4.4 GB137 memory-fit · 125 usable

Llama 3.2 3B at Q4_K_M on 140 phones

This table is fixed to the recommended Q4_K_M, on each phone's largest RAM option. Counts for smaller quants above will differ. All phones are shown, sorted by fit and estimated experience; tap one for the full report.

PhoneChip · RAMNeedsSpeedVerdict

~ = bandwidth-based estimate · ✓ = measured on real hardware

Llama 3.2 3B on 54 Mac configurations

Each Mac uses the recommended Q4_K_M at 4K context when it fits, or the largest smaller tracked quant that fits. Each row links to the full report with a macOS-specific setup guide.

MacUnified memoryQuantWorking budgetNeedsEstimated speedVerdict

FAQ

How much RAM do I need to run Llama 3.2 3B on a phone?

At Q4_K_M, Llama 3.2 3B needs ~2.9GB of usable memory (weights + KV cache + runtime). In practice that means a 6GB+ Android phone or a 6GB+ iPhone — iOS lets a single app use less of its RAM than Android does.

How fast does Llama 3.2 3B run on a flagship phone?

On the Galaxy S25 Ultra (Snapdragon 8 Elite) it runs at ~17.3 tokens/s at Q4_K_M, estimated from memory bandwidth. Anything above ~8 tokens/s feels smooth for chat.

Can an iPhone run Llama 3.2 3B?

Yes — the iPhone 17 runs it at ~15.3 tokens/s at Q4_K_M, using 2.9GB of its ~5.2GB usable memory.

What is the best quantization of Llama 3.2 3B for mobile?

Q4_K_M (2GB download) is the size/quality sweet spot of the 7 quants available. Total memory needed is ~2.9GB once the KV cache and runtime are counted — the download size alone understates it.

How many phones can run Llama 3.2 3B?

139 of the 140 phones we track have enough estimated usable memory at Q4_K_M; 139 also reach our usable-speed threshold of 3 tokens/s, and 103 run smoothly. Memory-fit alone does not mean a good experience.

Which Macs can run Llama 3.2 3B?

54 of the 54 exact Mac configurations we track have enough estimated working memory. Each Mac uses the recommended Q4_K_M when it fits, or the largest smaller tracked quant that fits, and separates installed unified memory, the macOS reserve, and estimated speed.

What is Llama 3.2 3B good for on a phone?

It's tagged for chat. At 3.2B parameters, it's a fast, lightweight pick for quick tasks.

Similar models

Llama 3.2 1BLlama 3.1 8BLlama 3.3 70BSmolLM3 3BLFM2.5-VL 3B