Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Mac

Re:Show HN: Open-source engine running Gemma 4 26B in 2 GB R... by znpy on Wednesday July 29, 2026 @11:33AM (#49098849)

I wonder if i can run this on my MacBook Neo!

[ Parent | on HN ]

Re:Show HN: Open-source engine running Gemma 4 26B in 2 GB R... by gitpusher42 on Wednesday July 29, 2026 @11:42AM (#49098975)

I haven't tried it but it should work! You can try it and share your results, it would be really appreciated

I tried it on my wife's M1 MacBook Air 512GB and it gets 4–5 tok/s

Also, it must be easy to adjust for iPhones and iPads in theory

[ Parent | on HN ]

Re:Show HN: Open-source engine running Gemma 4 26B in 2 GB R... by trollbridge on Wednesday July 29, 2026 @01:12PM (#49100242)

iPhones and iPads have much slower flash.

[ Parent | on HN ]

Re:Show HN: Open-source engine running Gemma 4 26B in 2 GB R... by jrgifford on Wednesday July 29, 2026 @09:43PM (#49105250)

Confirmed on my Neo!

Got 4.5 tokens / second sustained.

[ Parent | on HN ]

Re:Show HN: Open-source engine running Gemma 4 26B in 2 GB R... by gitpusher42 on Thursday July 30, 2026 @04:21AM (#49107318)

Thank you for testing and sharing, it is useful info!

[ Parent | on HN ]

Re:Show HN: Open-source engine running Gemma 4 26B in 2 GB R... by raver1975 on Thursday July 30, 2026 @03:24PM (#49114509)

I'm only getting < 1 token/second on my Neo, did you change some settings?

[ Parent | on HN ]

Re:Show HN: Open-source engine running Gemma 4 26B in 2 GB R... by gitpusher42 on Sunday August 02, 2026 @03:21AM (#49141950)

It is only a wild guess, but if you have 256gb version and a lot of apps running it can be pretty slow.

[ Parent | on HN ]

Re:Show HN: Open-source engine running Gemma 4 26B in 2 GB R... by trollbridge on Wednesday July 29, 2026 @01:12PM (#49100239)

Yes, a Neo is roughly equivalent to an M1.

[ Parent | on HN ]