You want a local AI model that runs well on your iPhone, and you do not want to compare quantization names to find it. Off Grid AI Mobile (OGAM) now opens the Models screen with three picks worked out for your iPhone's memory: Fast, Balanced and Best. Tap one, it downloads, and it becomes the model your chats use. After the download, the model runs on your iPhone, on the GPU through Metal, and nothing you type leaves the phone.
The same three picks sit at the top of the Image tab for image generation and the Speech tab for voice transcription.
What you need
- An iPhone on iOS 17 or later. The picks scale with RAM, so an iPhone with 6 GB sees smaller models than an iPhone 17 Pro Max with 12 GB.
- OGAM 0.0.117 or later from the App Store. That App Store release, out on 2026-10-23, brought the picks to iPhone. Download OGAM for iPhone.
- Wi-Fi for the download. Chat models range from a few hundred megabytes to a few gigabytes. After that, no connection is needed.
- No account and no payment. The picks are part of the free app.
How OGAM chooses Fast, Balanced and Best
The picks come from one rule in the app's model catalog, based on your iPhone's total RAM and each model's download size.
iOS does not give one app all of the phone's memory, and a running model needs more than its file size: the weights, the cache that holds your conversation, and working buffers. The team measured this on a 12 GB iPhone 17 Pro Max. The app had 6.0 GB free, and Gemma 4 E4B, a 4.64 GB download, needed 6.4 GB at a 4K context, about 1.4 times its file size. From that measurement the app sets the phone budget:
- A file under about 36% of your RAM runs well.
- A file between 36% and 50% of your RAM is marked Tight.
- A larger file is not offered.
| iPhone RAM | Runs well | Tight, up to |
|---|---|---|
| 6 GB | under 2.2 GB | 3 GB |
| 8 GB | under 2.9 GB | 4 GB |
| 12 GB | under 4.3 GB | 6 GB |
From the models that run well:
- Best is the highest-quality chat model in the catalog that fits.
- Balanced is the next best.
- Fast runs at the speed of a 4B model or faster, and faster than Balanced. A mixture-of-experts model counts by its active parameters, so a large model with a small active part can be the Fast pick.
Picks come from makers people know, such as Qwen, Gemma, Llama, Mistral and Phi. A model that a newer one replaces is never picked. The release notes give one example: a 12 GB phone gets Qwen 3.5 0.8B, 2B and 4B.
Step 1: Open the picks
- Open OGAM and tap Models in the tab bar at the bottom.
- Stay on the Text tab. At the top, under For this phone, three small cards sit side by side: Fast, Balanced and Best.
- Each card shows the model name and its download size.
Step 2: Download one and switch to it
- Tap the download icon on the card you want.
- When the download finishes, the model becomes your active chat model. The card gets a highlighted border and a chat icon.
- Tap the chat icon (Chat now) to open a chat with that model.
If you downloaded a pick earlier, its card shows a circle icon instead. Tap it to switch to that model with no new download.
A model you download from the list below the picks also becomes your active model when it finishes. The app shows "[model name] is ready" with Later and Chat now.
If you switch to a model that does not fit in the memory free right now, the app keeps your current model loaded, so the chat you are in keeps working.
Step 3: Pick an image model and a speech model
- Tap the Image tab. The same row of cards appears under For this phone, in place of the old RAM banner.
- On iPhone, the image models are Core ML builds of Stable Diffusion that run on the Apple Neural Engine. Variants of one model, such as a low-RAM build, count as one pick, and the card offers the variant suited to your iPhone.
- Tap the download icon on a card. When it finishes, the pick becomes the image model your chats use.
- Tap the Speech tab. The picks here are multilingual speech-to-text models only, so they work in whatever language you speak.
Image and speech models have no quality score in the catalog, so these picks go by size: Best is the largest that runs well, Fast the smallest of the rest, and Balanced the one nearest the middle.
Step 4: Choose a size in plain words
Tap a model in the list to open its details. The Sizes section shows up to three choices instead of quantization names:
- Recommended: the file that suits your iPhone best.
- Smaller: the smallest file below the recommended one, for when storage or memory is short.
- Best: the largest file above the recommended one, for the highest quality your iPhone can run.
Every file is still there. Tap All files under the list to see them all, and Fewer sizes to go back. Files too big for your iPhone are hidden, and files that only just fit carry a Tight label.
Step 5: Get new models without an app update
New models reach the catalog between App Store releases. To get them:
- On the Models screen, tap the refresh icon at the top right, next to the downloads icon.
- A sheet opens and names each new model, for example "2 new models". The new models show in the list straight away.
- If nothing changed, the sheet says Up to date. If your iPhone is offline, it says it could not check and asks you to check your connection.
The picks are worked out again with the new models included, so a better model can move into the Best slot.
What this does not do
- It does not run a speed test. The picks come from RAM, file size and the catalog's quality ranking. To tune speed for the model you picked, use Use fastest settings in the text model settings. It tests each engine your iPhone can run in 30 seconds or less, and you can cancel it.
- It does not check free storage when it picks. The app checks free space when the download starts and tells you how much it needs if space is short.
- It does not offer SDXL on iPhone. Its Core ML build needs about 7 GB of memory while it runs, and iOS closes the app before it finishes, even on the newest iPhones.
- It does not pick from all of Hugging Face. The picks come from the curated catalog. You can still search Hugging Face from the Text tab, and the search hides files your iPhone cannot run.
- A slot can be empty. On an iPhone with less RAM, or on the shorter image list, the row shows only the slots it can fill.
FAQ
What is the best local AI model for an iPhone with 8 GB of RAM?
Open the Text tab in OGAM and look at the Best card: that is the highest-quality catalog model that runs well in 8 GB. On 8 GB, files under about 2.9 GB run well, so the picks are small models in 4-bit builds.
Which is faster on iPhone, Fast or Balanced?
Fast. It runs at the speed of a 4B model or faster, while Balanced is the next-best quality model and can be larger.
Does picking a model need internet?
Only for the download and for checking for new models. After the download, chat, image generation and transcription with that model run on the iPhone with no connection.
Why does a model show Tight?
Its file is between about 36% and 50% of your iPhone's RAM. It can load, but there is little room left for other apps and a long conversation, and iOS may close the app under memory pressure. A smaller size is steadier.
Can I still choose a different model?
Yes. The list below the picks has the full curated catalog, and you can search Hugging Face from the same tab.
Top comments (0)