text model · DeepSeek-R1-Distill · iOS
Can I run DeepSeek-R1-0528-Qwen3-8B on iPad Pro M4 (16GB, 1TB/2TB config)?
Yes. DeepSeek-R1-0528-Qwen3-8B runs on iPad Pro M4 (16GB, 1TB/2TB config) at Q4_K_M (~6.2 GB of ~12 GB usable).
Runs at Q4_K_M using ~6.2 GB of ~12 GB usable. You have room for Q8_0 for higher quality.
That figure is at a 4k context and moves about ±15% as context length changes. iPad Pro M4 (16GB, 1TB/2TB config) leaves ~5.8 GB of headroom, room to step up to Q8_0 for higher quality.
- Q4_K_M needed
- ~6.2 GB
- Usable on device
- ~12 GB
- Device memory
- 16 GB
- Best quant
- Q4_K_M
Which quant fits
Running cost · estimate
- Power draw
- ~14 W
- Electricity / 1M tokens
- ~$0.04
- Pays for itself after
- ~3,476M tok
At ~$0.15/kWh and the estimated ~13 tok/s, a million generated tokens costs about $0.04 in electricity. Versus a hosted API at ~$0.5 per million tokens, the ~$1,599 iPad Pro M4 (16GB, 1TB/2TB config) pays for itself after roughly 3,476 million tokens, so local hardware is mostly a fixed cost, not a per-token one. TDP is the peak draw, so this is an upper bound. Assumptions.
How to run it
On iOS use Apple Foundation Models (Built into iOS 26, ~3B on-device model, zero download, fully private.).
- Parameters
- 8.19B
- Q4_K_M size
- 4.68 GB
- Q8_0 size
- 8.11 GB
- Context
- 128k
- Ollama tag
- deepseek-r1:8b-0528-qwen3-q4_K_M
- Memory
- 16 GB unified
- Usable for weights
- ~12 GB
- Power draw
- ~14 W
- Best runtime
- MLX (via Python or Swift; mlx-lm package)
You could also run
Run DeepSeek-R1-0528-Qwen3-8B on other hardware
FAQ
Can iPad Pro M4 (16GB, 1TB/2TB config) run DeepSeek-R1-0528-Qwen3-8B?
Yes. DeepSeek-R1-0528-Qwen3-8B runs on iPad Pro M4 (16GB, 1TB/2TB config) at Q4_K_M (~6.2 GB of ~12 GB usable).
How much memory does DeepSeek-R1-0528-Qwen3-8B need?
iPad Pro M4 (16GB, 1TB/2TB config) has room to spare. At Q4_K_M the weights are ~4.68 GB; with KV cache and runtime overhead, budget ~6.2 GB at a 4k context.
What is the best tool to run DeepSeek-R1-0528-Qwen3-8B 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/deepseek-r1-0528-qwen3-8b/ipad-pro-m4-16gb) Sources
- apple.com/ipad-pro
- apple.com/newsroom
- developer.apple.com
- enclaveai.app
- github.com/a-ghorbani
- github.com/google-ai-edge
- github.com/mainframecomputer
- gsmarena.com
- huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
- huggingface.co/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B/blob
- huggingface.co/unsloth
- layla-network.ai
- ollama.com
- phonearena.com
- privatellm.app
- support.apple.com/en-us/119891
- support.apple.com/en-us/119892
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.