AREX Turbo 4B

At Q4_K_M, 132 of 135 phones have enough estimated usable memory; 122 also reach 3+ tokens/s. On Mac, 38 of 38 exact configurations pass the working-memory check.

Params
4.5B
Family
arex
Released
2026-07
Tags
chat · reasoning

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
Q2_K2.1 GB~3.1 GB134 memory-fit · 134 usable
Q3_K_M2.4 GB~3.4 GB134 memory-fit · 134 usable
IQ4_XS2.5 GB~3.5 GB134 memory-fit · 134 usable
Q4_02.6 GB~3.6 GB134 memory-fit · 134 usable
Q4_K_S2.7 GB~3.8 GB134 memory-fit · 124 usable
Q4_K_M2.9 GB~4 GB132 memory-fit · 122 usable
Q5_K_M3.3 GB~4.4 GB132 memory-fit · 122 usable
Q6_K3.7 GB~4.8 GB132 memory-fit · 122 usable
Q8_04.5 GB~5.6 GB125 memory-fit · 94 usable

AREX Turbo 4B at Q4_K_M on 135 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

AREX Turbo 4B on 38 Mac configurations

Fixed to Q4_K_M at 4K context. Each row links to the same full `/run` report used for phones, with a macOS-specific setup guide.

MacUnified memoryWorking budgetNeedsEstimated speedVerdict

FAQ

How much RAM do I need to run AREX Turbo 4B on a phone?

At Q4_K_M, AREX Turbo 4B needs ~4GB of usable memory (weights + KV cache + runtime). In practice that means a 6GB+ Android phone or a 8GB+ iPhone — iOS lets a single app use less of its RAM than Android does.

How fast does AREX Turbo 4B run on a flagship phone?

On the OPPO Find X8 Pro (Dimensity 9400) it runs at ~13.2 tokens/s at Q4_K_M, estimated from memory bandwidth. Anything above ~8 tokens/s feels smooth for chat.

Can an iPhone run AREX Turbo 4B?

Yes — the iPhone Air runs it at ~11.9 tokens/s at Q4_K_M, using 4GB of its ~7.8GB usable memory.

What is the best quantization of AREX Turbo 4B for mobile?

Q4_K_M (2.9GB download) is the size/quality sweet spot of the 9 quants available. Total memory needed is ~4GB once the KV cache and runtime are counted — the download size alone understates it.

How many phones can run AREX Turbo 4B?

132 of the 135 phones we track have enough estimated usable memory at Q4_K_M; 122 also reach our usable-speed threshold of 3 tokens/s, and 95 run smoothly. Memory-fit alone does not mean a good experience.

Which Macs can run AREX Turbo 4B?

38 of the 38 exact Mac configurations we track have enough estimated working memory at Q4_K_M. Each Mac result separates installed unified memory, the macOS reserve, estimated speed, and quant choices.

What is AREX Turbo 4B good for on a phone?

It's tagged for chat, reasoning. At 4.5B parameters, it balances quality and speed for daily use.

Similar models

Fara 1.5 4BGemma 3 4BQwen3 4BQwen 3.5 4BGemma 4 E4B