text model · Olmo · iOS
Can I run Olmo 3.1 32B Instruct on iPhone Air?
No. Olmo 3.1 32B Instruct needs ~20.3 GB even at Q4_K_M, but iPhone Air only has ~8 GB usable.
Needs ~20.3 GB even at Q4_K_M, but only ~8 GB is usable.
That figure is at a 4k context and moves about ±15% as context length changes.
The gap is about 12.3 GB: Olmo 3.1 32B Instruct needs roughly 20.3 GB at Q4_K_M and iPhone Air leaves only about 8 GB usable for a model. The lightest tracked hardware that runs Olmo 3.1 32B Instruct is the Nvidia GeForce RTX 4090 (24GB) at 24 GB. See Olmo 3.1 32B Instruct on Nvidia GeForce RTX 4090 (24GB).
- Q4_K_M needed
- ~20.3 GB
- Usable on device
- ~8 GB
- Device memory
- 12 GB
Which quant fits
How to run it
On iOS use Apple Foundation Models (Built into iOS 26, ~3B on-device model, zero download, fully private.).
- Parameters
- 32B
- Q4_K_M size
- 18.14 GB
- Q8_0 size
- 31.9 GB
- Context
- 64k
- Ollama tag
- olmo-3.1:32b-instruct
- Memory
- 12 GB unified
- Usable for weights
- ~8 GB
- Best runtime
- llama.cpp + Metal (via PocketPal or Off Grid app)
What you can run instead
Run Olmo 3.1 32B Instruct on other hardware
FAQ
Can iPhone Air run Olmo 3.1 32B Instruct?
No. Olmo 3.1 32B Instruct needs ~20.3 GB even at Q4_K_M, but iPhone Air only has ~8 GB usable.
How much memory does Olmo 3.1 32B Instruct need?
iPhone Air does not have enough memory. At Q4_K_M the weights are ~18.14 GB; with KV cache and runtime overhead, budget ~20.3 GB at a 4k context.
What is the best tool to run Olmo 3.1 32B Instruct on iOS?
On iPhone and iPad, Apple Foundation Models (Built into iOS 26, ~3B on-device model, zero download, fully private.) is the standard choice. Phones realistically run 1B-4B class models. Anything larger thermally throttles or OOMs.
Embed this
A live badge for your model card or README, updated as the data is. [](https://localmodel.run/can-i-run/olmo-3.1-32b-instruct/iphone-air) Sources
Weights are measured from GGUF files; KV cache and overhead are computed, so totals can vary ~15% with context and runtime. Any tok/s is a bandwidth estimate. See methodology.