aicanrun

Can OneJev 27B run on MacBook Air M1 16GB?

Weights exceed the memory budget · custom workflow
Q4_K_M · 4K context · formula estimate

What this means

The checked Q4_K_M build needs about 20GB, above this Mac's conservative 13GB working budget.

OneJev scores typed decisions in one forward pass. Generic decode estimates are not decision latency; image inputs need additional projector memory. Use the publisher's qev server and System One client.

View technical details ↓

Not measured
decision latency
20GB
needed at Q4_K_M
Q4_K_M
quant selected
0GB
working headroom
needs 20 GBworking budget 13 GB · short 7 GB
Apple M116 GB unified memoryMetalFanless

Where the memory goes

ComponentDetailGB
Model weightsQ4_K_M GGUF (16.5 GB) + mmap overhead17.3
KV cache4K context window1.9
RuntimemacOS inference app + compute buffers0.8
Total neededat Q4_K_M, 4K context20
Working budget16 GB unified memory − conservative macOS reserve13
Headroommemory shortfall−7

Pick your quant

QuantDownloadMemoryEstimated speedVerdict
Q2_K10.7 GB13.9 GB—✕ Won't fit
Q3_K_M13.3 GB16.7 GB—✕ Won't fit
IQ4_XS15.2 GB18.7 GB—✕ Won't fit
Q4_K_S15.6 GB19.1 GB—✕ Won't fit
Q4_K_M ★16.5 GB20 GB—✕ Won't fit
Q5_K_M19.2 GB22.9 GB—✕ Won't fit
Q6_K22.1 GB25.9 GB—✕ Won't fit
Q8_028.6 GB32.7 GB—✕ Won't fit

Other models on this Mac

Bonsai 27B (1-bit) · Q1_0Ternary Bonsai 27B · Q2_0Qwen3 14B · Q4_K_MPhi-4 14B · Q4_K_MMinistral 3 14B · Q4_K_MGemma 3 12B · Q4_K_M

OneJev 27B on other MacBook Air M1 configurations

Check 8GB

FAQ

Can the MacBook Air M1 · 16GB run OneJev 27B?

The tracked Q4_K_M working memory is estimated at 20GB, excluding the vision projector. Weight fit is not verified workflow support; use the publisher's qev server and System One client.

Which OneJev 27B quant should I use on this Mac?

None of the tracked quants fit this configuration safely. Choose a smaller model or a Mac with more unified memory.

Which app should I use for OneJev 27B on this Mac?

OneJev 27B is a task-specific model, not a normal local chat download. The selected weights do not fit this Mac, and the model also needs its publisher's intended workflow. This page links that repository and recommends Llama 3.2 3B if you just want local chat.

Are these speeds measured on a MacBook Air M1?

Decision latency has not been measured. OneJev scores typed choices in one forward pass; generic decode estimates do not measure this workflow.