text model · Qwen3 · iOS
Can I run Qwen3 235B A22B on iPhone 16 Pro?
No. Qwen3 235B A22B needs ~136.9 GB even at Q4_K_M, but iPhone 16 Pro only has ~4.5 GB usable.
Needs ~136.9 GB even at Q4_K_M, but only ~4.5 GB is usable.
That figure is at a 4k context and moves about ±15% as context length changes.
The gap is about 132.4 GB: Qwen3 235B A22B needs roughly 136.9 GB at Q4_K_M and iPhone 16 Pro leaves only about 4.5 GB usable for a model. The lightest tracked hardware that runs Qwen3 235B A22B is the Apple M3 Ultra (256GB) at 256 GB. See Qwen3 235B A22B on Apple M3 Ultra (256GB).
- Q4_K_M needed
- ~136.9 GB
- Usable on device
- ~4.5 GB
- Device memory
- 8 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
- 235B (MoE, 22B active)
- Q4_K_M size
- 132.39 GB
- Q8_0 size
- 232.77 GB
- Context
- 128k
- Ollama tag
- qwen3:235b
- Memory
- 8 GB unified
- Usable for weights
- ~4.5 GB
- Power draw
- ~12 W
- Best runtime
- llama.cpp + Metal (via PocketPal or Off Grid app)
What you can run instead
Run Qwen3 235B A22B on other hardware
FAQ
Can iPhone 16 Pro run Qwen3 235B A22B?
No. Qwen3 235B A22B needs ~136.9 GB even at Q4_K_M, but iPhone 16 Pro only has ~4.5 GB usable.
How much memory does Qwen3 235B A22B need?
iPhone 16 Pro does not have enough memory. At Q4_K_M the weights are ~132.39 GB; with KV cache and runtime overhead, budget ~136.9 GB at a 4k context. It is a Mixture-of-Experts model (235B total / 22B active), so all experts must stay in memory; memory tracks total params, not active params.
What is the best tool to run Qwen3 235B A22B 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/qwen3-235b-a22b/iphone-16-pro) Sources
- abachy.com
- aider.chat
- apple.com/iphone-16-pro
- apple.com/newsroom
- developer.apple.com
- en.wikipedia.org
- enclaveai.app
- github.com/a-ghorbani
- github.com/google-ai-edge
- github.com/mainframecomputer
- gsmarena.com
- huggingface.co/Qwen
- huggingface.co/unsloth
- layla-network.ai
- macrumors.com
- ollama.com/library/qwen3
- ollama.com/library/qwen3/tags
- privatellm.app
- versus.com
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.