Gemma 3 12B

At the default Q4_K_M, 67 of 151 phones have enough estimated usable memory; 67 also reach 3+ tokens/s. On Mac, 54 of 54 exact configurations pass using that quant or a smaller tracked fallback.

Params
12.2B
Family
gemma
Released
2025-03
Tags
chat · vision

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
IQ1_S3.1 GB~4.7 GB146 memory-fit · 134 usable
Q2_K4.8 GB~6.5 GB114 memory-fit · 100 usable
Q3_K_M6 GB~7.8 GB114 memory-fit · 98 usable
IQ4_XS6.6 GB~8.4 GB111 memory-fit · 95 usable
Q4_06.9 GB~8.7 GB111 memory-fit · 95 usable
Q4_K_S6.9 GB~8.7 GB111 memory-fit · 95 usable
Q4_K_M ★7.3 GB~9.1 GB67 memory-fit · 67 usable
Q5_K_M8.4 GB~10.3 GB67 memory-fit · 67 usable
Q6_K9.7 GB~11.6 GB67 memory-fit · 67 usable
Q8_012.5 GB~14.6 GB4 memory-fit · 0 usable

Gemma 3 12B at Q4_K_M on 151 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

Gemma 3 12B 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 Gemma 3 12B on a phone?

At Q4_K_M, Gemma 3 12B needs ~9.1GB of usable memory (weights + KV cache + runtime). In practice that means a 16GB+ Android phone or a 16GB+ iPhone — iOS lets a single app use less of its RAM than Android does.

How fast does Gemma 3 12B run on a flagship phone?

On the Xiaomi 18 Fold (Xiaomi XRING O3) it runs at ~7 tokens/s at Q4_K_M, estimated from memory bandwidth. Anything above ~8 tokens/s feels smooth for chat.

Can an iPhone run Gemma 3 12B?

Not really: the best iPhone we track (iPhone Air) has ~7.8GB usable, but Gemma 3 12B needs 9.1GB at Q4_K_M.

What is the best quantization of Gemma 3 12B for mobile?

Q4_K_M (7.3GB download) is the size/quality sweet spot of the 10 quants available. Total memory needed is ~9.1GB once the KV cache and runtime are counted — the download size alone understates it.

How many phones can run Gemma 3 12B?

67 of the 151 phones we track have enough estimated usable memory at Q4_K_M; 67 also reach our usable-speed threshold of 3 tokens/s, and 0 run smoothly. Memory-fit alone does not mean a good experience.

Which Macs can run Gemma 3 12B?

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 Gemma 3 12B good for on a phone?

It's tagged for chat, vision. At 12.2B parameters, it prioritizes answer quality over speed — expect slower decoding.

Similar models

Gemma 3 1BGemma 3 4BGemma 4 E2BGemma 4 E4BGemma 4 12B