Llama 3.2 3B

Needs ~2.9GB of memory at Q4_K_M and runs on 98 of the 99 phones we track — 67 of them smoothly.

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. ★ = recommended.

QuantDownloadMemory neededFits on
IQ4_XS1.8 GB~2.7 GB98 of 99 phones
Q4_01.9 GB~2.8 GB98 of 99 phones
Q4_K_S1.9 GB~2.8 GB98 of 99 phones
Q4_K_M2 GB~2.9 GB98 of 99 phones
Q5_K_M2.3 GB~3.2 GB98 of 99 phones
Q6_K2.6 GB~3.6 GB98 of 99 phones
Q8_03.4 GB~4.4 GB96 of 99 phones

Llama 3.2 3B on 99 phones

At Q4_K_M, on each phone's largest RAM option. Tap a phone for the full report.

PhoneChip · RAMNeedsSpeedVerdict
SGalaxy S25 UltraSnapdragon 8 Elite · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SGalaxy S25Snapdragon 8 Elite · 12 GB2.9 / 9 GB~17.3 tokens/s Runs great
SGalaxy S24 UltraSnapdragon 8 Gen 3 · 12 GB2.9 / 9 GB~17.3 tokens/s Runs great
SGalaxy S24Samsung Exynos 2400 · 12 GB2.9 / 9 GB~15.1 tokens/s Runs great
SGalaxy S23 UltraSnapdragon 8 Gen 2 · 12 GB2.9 / 9 GB~15.1 tokens/s Runs great
SPixel 9 ProGoogle Tensor G4 · 16 GB2.9 / 12 GB~15.4 tokens/s Runs great
SPixel 9Google Tensor G4 · 12 GB2.9 / 9 GB~15.4 tokens/s Runs great
SPixel 8 ProGoogle Tensor G3 · 12 GB2.9 / 9 GB~15.4 tokens/s Runs great
SPixel 8aGoogle Tensor G3 · 8 GB2.9 / 6 GB~15.4 tokens/s Runs great
SOnePlus 13Snapdragon 8 Elite · 24 GB2.9 / 20 GB~17.3 tokens/s Runs great
SOnePlus 12Snapdragon 8 Gen 3 · 24 GB2.9 / 20 GB~17.3 tokens/s Runs great
SXiaomi 15 ProSnapdragon 8 Elite · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SXiaomi 14Snapdragon 8 Gen 3 · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SPOCO X6 ProDimensity 8300 Ultra · 12 GB2.9 / 9 GB~15.3 tokens/s Runs great
SOPPO Find X8 ProDimensity 9400 · 16 GB2.9 / 12 GB~19.2 tokens/s Runs great
Svivo X200 ProDimensity 9400 · 16 GB2.9 / 12 GB~19.2 tokens/s Runs great
Svivo X100 ProDimensity 9300 · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SROG Phone 9 ProSnapdragon 8 Elite · 24 GB2.9 / 20 GB~17.3 tokens/s Runs great
SNothing Phone (3)Snapdragon 8s Gen 4 · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SMotorola Edge 50 UltraSnapdragon 8s Gen 3 · 16 GB2.9 / 12 GB~15.1 tokens/s Runs great
SHONOR Magic7 ProSnapdragon 8 Elite · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SiPhone 17Apple A19 · 8 GB2.9 / 5.2 GB~15.3 tokens/s Runs great
SiPhone AirApple A19 Pro · 12 GB2.9 / 7.8 GB~17.3 tokens/s Runs great
SiPhone 17 ProApple A19 Pro · 12 GB2.9 / 7.8 GB~17.3 tokens/s Runs great
SiPhone 17 Pro MaxApple A19 Pro · 12 GB2.9 / 7.8 GB~17.3 tokens/s Runs great
SGalaxy S26Snapdragon 8 Elite Gen 5 · 12 GB2.9 / 9 GB~19.1 tokens/s Runs great
SGalaxy S26+Snapdragon 8 Elite Gen 5 · 12 GB2.9 / 9 GB~19.1 tokens/s Runs great
SGalaxy S26 UltraSnapdragon 8 Elite Gen 5 · 16 GB2.9 / 12 GB~19.1 tokens/s Runs great
SXiaomi 17Snapdragon 8 Elite Gen 5 · 16 GB2.9 / 12 GB~19.1 tokens/s Runs great
SXiaomi 17 ProSnapdragon 8 Elite Gen 5 · 16 GB2.9 / 12 GB~19.1 tokens/s Runs great
SOnePlus 15Snapdragon 8 Elite Gen 5 · 16 GB2.9 / 12 GB~19.1 tokens/s Runs great
Svivo X300 ProDimensity 9500 · 16 GB2.9 / 12 GB~19.2 tokens/s Runs great
SOPPO Find X9 ProDimensity 9500 · 16 GB2.9 / 12 GB~19.2 tokens/s Runs great
SPixel 10Google Tensor G5 · 12 GB2.9 / 9 GB~15.3 tokens/s Runs great
SPixel 10 ProGoogle Tensor G5 · 16 GB2.9 / 12 GB~15.3 tokens/s Runs great
SPixel 9aGoogle Tensor G4 · 8 GB2.9 / 6 GB~15.4 tokens/s Runs great
SGalaxy S25 FESamsung Exynos 2400 · 8 GB2.9 / 6 GB~15.1 tokens/s Runs great
SGalaxy Z Fold6Snapdragon 8 Gen 3 · 12 GB2.9 / 9 GB~17.3 tokens/s Runs great
SGalaxy Z Flip6Snapdragon 8 Gen 3 · 12 GB2.9 / 9 GB~17.3 tokens/s Runs great
SGalaxy Z Fold7Snapdragon 8 Elite · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SGalaxy Z Flip7Exynos 2500 · 12 GB2.9 / 9 GB~17.3 tokens/s Runs great
SMotorola Razr Ultra (2025)Snapdragon 8 Elite · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SOnePlus 13RSnapdragon 8 Gen 3 · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SOnePlus 15RSnapdragon 8 Gen 5 · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SPixel 9 Pro XLGoogle Tensor G4 · 16 GB2.9 / 12 GB~15.4 tokens/s Runs great
SPixel 10 Pro XLGoogle Tensor G5 · 16 GB2.9 / 12 GB~15.3 tokens/s Runs great
SiPhone 17eApple A19 · 8 GB2.9 / 5.2 GB~15.3 tokens/s Runs great
SPixel 10aGoogle Tensor G4 · 8 GB2.9 / 6 GB~15.4 tokens/s Runs great
SMotorola Razr+ (2026)Snapdragon 8s Gen 3 · 12 GB2.9 / 9 GB~15.1 tokens/s Runs great
SMotorola Razr Ultra (2026)Snapdragon 8 Elite · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SRedMagic 11 AirSnapdragon 8 Elite · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SRedMagic 11S ProSnapdragon 8 Elite Gen 5 · 16 GB2.9 / 12 GB~19.1 tokens/s Runs great
SXiaomi 15TDimensity 8400 Ultra · 12 GB2.9 / 9 GB~15.4 tokens/s Runs great
SXiaomi 15T ProDimensity 9400+ · 12 GB2.9 / 9 GB~19.2 tokens/s Runs great
SPOCO X7 ProDimensity 8400 Ultra · 12 GB2.9 / 9 GB~15.4 tokens/s Runs great
SHonor Magic8 ProSnapdragon 8 Elite Gen 5 · 16 GB2.9 / 12 GB~19.1 tokens/s Runs great
SAQUOS R10Snapdragon 7+ Gen 3 · 12 GB2.9 / 9 GB~15.1 tokens/s Runs great
SXperia 1 VIISnapdragon 8 Elite · 16 GB2.9 / 12 GB~17.3 tokens/s Runs great
SXperia 1 VIIISnapdragon 8 Elite Gen 5 · 16 GB2.9 / 12 GB~19.1 tokens/s Runs great
SGalaxy Z Fold8Snapdragon 8 Elite Gen 5 · 16 GB2.9 / 12 GB~19.1 tokens/s Runs great
SGalaxy Z Flip8Snapdragon 8 Elite Gen 5 · 12 GB2.9 / 9 GB~19.1 tokens/s Runs great
SiPhone 16 Pro MaxApple A18 Pro · 8 GB2.9 / 5.2 GB~13.5 tokens/s Runs great
SiPhone 16 ProApple A18 Pro · 8 GB2.9 / 5.2 GB~13.5 tokens/s Runs great
SiPhone 16Apple A18 · 8 GB2.9 / 5.2 GB~13.5 tokens/s Runs great
SiPhone 16eApple A18 · 8 GB2.9 / 5.2 GB~13.5 tokens/s Runs great
SiPhone 15 ProApple A17 Pro · 8 GB2.9 / 5.2 GB~11.5 tokens/s Runs great
AGalaxy A57 5GExynos 1680 · 12 GB2.9 / 9 GB~7.7 tokens/s! Runs, barely
AiPhone 15Apple A16 Bionic · 6 GB2.9 / 3.9 GB~11.5 tokens/s Runs great
ANothing Phone (4a) ProSnapdragon 7 Gen 4 · 12 GB2.9 / 9 GB~7.6 tokens/s! Runs, barely
AGalaxy A55Exynos 1480 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
ARedmi Note 13 ProSnapdragon 7s Gen 2 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
ANothing Phone (2a)Dimensity 7200 Pro · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
ANothing Phone (3a)Snapdragon 7s Gen 3 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
AGalaxy A26 5GExynos 1380 · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AGalaxy A36 5GSnapdragon 6 Gen 3 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
AGalaxy A56 5GExynos 1580 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
AMoto G Stylus (2025)Snapdragon 6 Gen 3 · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AMoto G Stylus (2026)Snapdragon 6 Gen 3 · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AMotorola Razr (2025)Dimensity 7400X · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AGalaxy A37 5GExynos 1480 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
AMotorola Razr (2026)Dimensity 7450X · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AMotorola Edge (2026)Dimensity 7450 · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
ARedmi Note 14 Pro 5GDimensity 7300 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
ARedmi Note 15 Pro 5GDimensity 7400 Ultra · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
ARedmi Note 15 5GSnapdragon 6 Gen 3 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
ARedmi Note 14 Pro+ 5GSnapdragon 7s Gen 3 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
ARedmi Note 15 Pro+ 5GSnapdragon 7s Gen 4 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
AHonor 400Snapdragon 7 Gen 3 · 12 GB2.9 / 9 GB~5.8 tokens/s! Runs, barely
ACMF Phone 2 ProDimensity 7300 · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AAQUOS sense10Snapdragon 7s Gen 3 · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AAQUOS sense9Snapdragon 7s Gen 2 · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AXperia 10 VIISnapdragon 6 Gen 3 · 8 GB2.9 / 6 GB~5.8 tokens/s! Runs, barely
AiPhone 14Apple A15 Bionic · 6 GB2.9 / 3.9 GB~7.7 tokens/s! Runs, barely
AGalaxy A16 5GExynos 1330 · 8 GB2.9 / 6 GB~3.8 tokens/s! Runs, barely
AGalaxy A17 5GExynos 1330 · 8 GB2.9 / 6 GB~3.8 tokens/s! Runs, barely
AMoto G (2025)Dimensity 6300 · 8 GB2.9 / 6 GB~3.8 tokens/s! Runs, barely
AMoto G Power (2025)Dimensity 6300 · 8 GB2.9 / 6 GB~3.8 tokens/s! Runs, barely
AMoto G Power (2026)Dimensity 6300 · 8 GB2.9 / 6 GB~3.8 tokens/s! Runs, barely
FMoto G (2026)Dimensity 6300 · 4 GB2.9 / 2 GB Won't fit

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

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?

98 of the 99 phones we track fit it at Q4_K_M; 67 run it smoothly. The rest fall short on usable RAM.

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 3BMinistral 3 3B