Home › Photo & Video
PY
USen-US
CNzh-Hans-CN
DEde-DE
JPja
RUru
GBen-GB
CAen-CA
TWzh-Hant-TW
FRfr-FR
ESes-ES
HKzh-Hant-HK
AUen-AU
THen-GB
INen-GB
ITit
KRko
NLnl
CHde-CH
PLen-GB
UAen-GB
MXes-MX
TRen-GB
SEsv
BRpt-BR
ATde-DE
IDen-GB
SGen-GB
ZAen-GB
FIen-GB
MYen-GB
VNen-GB
CZen-GB
PHen-GB
PTpt-PT
BEen-GB
BYen-GB
HUen-GB
KZen-GB
NZen-AU
ROen-GB
DKen-GB
NOen-GB
AEen-GB
EGen-GB
LUen-GB
EEen-GB
IQen-GB
KGen-GB
LKen-GB
LTen-GB
MDen-GB
NPen-GB
PEes-MX
PKen-GB
RSen-GB
SIen-GB
AZen-GB
BGen-GB
BNen-GB
CRes-MX
DOes-MX
GEen-GB
GTes-MX
HRen-GB
IEen-GB
ISen-GB
KWen-GB
LVen-GB
MAen-GB
MKen-GB
MTen-GB
PAes-MX
QAen-GB
SKen-GB
SVes-MX
TTen-GB
AFen-GB
AGen-GB
AIen-GB
ALen-GB
AMen-GB
AOen-GB
BAen-GB
BBen-GB
BFen-GB
BJen-GB
BMen-GB
BOes-MX
BSen-GB
BTen-GB
BWen-GB
BZen-GB
CDen-GB
CGen-GB
CIfr-FR
CMfr-FR
CVen-GB
DMen-GB
DZen-GB
FJen-GB
FMen-GB
GAfr-FR
GDen-GB
GHen-GB
GMen-GB
GRen-GB
GWen-GB
GYen-GB
HNes-MX
JMen-GB
KEen-GB
KHen-GB
KNen-GB
KYen-GB
LAen-GB
LCen-GB
LRen-GB
LYen-GB
MEen-GB
MGen-GB
MLen-GB
MMen-GB
MNen-GB
MOzh-Hant-HK
MRen-GB
MSen-GB
MUen-GB
MVen-GB
MZen-GB
NAen-GB
NEen-GB
NGen-GB
NIes-MX
NRen-GB
OMen-GB
PGen-GB
PWen-GBPYes-MX
RWen-GB
SBen-GB
SCen-GB
SLen-GB
SNen-GB
SRen-GB
STen-GB
SZen-GB
TCen-GB
TDen-GB
TJen-GB
TMen-GB
TNen-GB
TOen-GB
TZen-GB
UGen-GB
UZen-GB
VCen-GB
VEes-MX
VGen-GB
VUen-GB
XKen-GB
YEen-GB
ZMen-GB
ZWen-GB
163 of 163 storefronts
Content in en-US
Content in en-US
Information
Apple ID6444050820
Bundle IDcom.liuliu.draw-things
SellerSeller
Min iOS15.4
Device familiesmac, iphone, ipad, ipod
Released2022-11-07
Copyright© 2026, Draw Things Inc.
Privacy policyLink ↗
Preview 3 platforms
iPhone 6 images
iPad 5 images
Mac 5 images
What’s New Version 1.20260518.2
* Add interrogator model "Qwen 3.5 4B".
* Add dedicated mode selector "Auto" / "Generation" / "Edit".
* Support "Sign in with Google".
* Some other small visual updates.
Description
Unleash your imagination with Draw Things, the powerful offline AI creation tool for your iPhone or iPad. Create stunning images privately, right on your device – no internet or subscriptions needed for local creation!
• Your Private AI Studio: Create art 100% offline. Your prompts and images stay securely on your device.
• Powerful AI Diffusion Engine: Harness the latest models like SDXL, FLUX.1, Wan 2.2, Qwen Image, Z Image, FLUX.2 and LTX-2.3 for incredible results.
• Pro-Level Control: Master image creation with on-device LoRA training, full ControlNet support, inpainting, outpainting, and pose editing.
• Unlimited Customization: Import your favorite community-made diffusion models and LoRAs effortlessly.
• Free Local Creation: Create limitlessly on your device without ads or token limits. (Optional Cloud Compute available).
• Optimized for Apple: Experience blazing-fast performance, especially on M1, M2, and M3 Apple Silicon Macs. Native support for iPhone, iPad, and Mac.
• Complete Creative Suite: Explore text-to-image, image-to-image, AI video creation, infinite canvas, scribble-to-image, and more.
Perfect for AI enthusiasts, artists, designers, and creators seeking unparalleled control over AI creation. New to AI image / video creation? Dive into our comprehensive Wiki and join our vibrant Discord community for guides, tips, and support!
EULA: https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
Privacy Policy: https://drawthings.ai/policy/
In-App Purchases PY pricing
Draw Things+ Subscription
Monthly—
One-Time Purchases
Boost Bundle S—
Boost Bundle M—
Boost Bundle L—
Capabilities
In-App Purchases
Accessibility
ipadiphonemacrealityDevicetvoswatch
Version history 25 versions
Build 885837368v1.20260518.22026-05-23 01:49:30
* Add interrogator model "Qwen 3.5 4B".
* Add dedicated mode selector "Auto" / "Generation" / "Edit".
* Support "Sign in with Google".
* Some other small visual updates.
Build 885219895v1.20260501.02026-05-04 15:13:53
* Hotfix for new segmented downloader issue.
Build 885140920v1.20260430.02026-05-01 17:47:59
* Support SeedVR2 as one of the diffusion model for 1-step upscale / restoration work.
* Fix Anima LLM adapter related issues.
* Switch to better newer causal / mask attention kernels.
* Support export LoRAs for Qwen, Z Image, FLUX.2 and ERNIE Image models.
* Imported Z Image models now runs at BF16 precision.
Build 884674149v1.20260418.12026-04-20 20:07:08
* Added support for ERNIE Image, including LoRA import, training and model import.
* Added support for Anima Preview 3, including LoRA import, training and model import.
* Updated LTX-2.3 [distilled] to LITX-2.3 [distilled] 1.1.
* Fixed crash related to Z-Image LoRA training.
* For M5 / A19 Pro, now defaults to ANE+NAX hybrid inference mode.
Build 884388056v1.20260410.12026-04-11 18:22:49
* Add 8-bit S support for M4 / A18 Pro devices, reaching up to 2x speed-up for these device.
* Support LoRA training for FLUX.2 [klein] 4B / 9B / Z Image / Qwen Image as base model.
* Fix audio sync issue during in-app playback.
* Fix some 8-bit S model generation issues on M5 / A19 Pro.
Build 883946199v1.20260330.02026-03-31 17:32:48
* Added a new quantization variant: 8-bit S, which on A19 Pro / M5 is about 1.5x to 2x faster than Exact or 8-bit variant.
* Fixed issue when importing models, it was no longer respecting External Folder selection.
* Support Bridge Mode to Cloud Compute with Bring Your Own LoRAs.
* Some language updates to our tiers, now it is Local Only, Free, Draw Things+.
Build 883685591v1.20260323.02026-03-24 17:43:42
* Support export video as HEVC;
* Fix new History item menu issue.
Build 883606787v1.20260320.12026-03-23 02:59:55
* Fix FLUX.2 [klein] 9B KV cannot use batch size or text guidance != 1 correctly.
* Better Hires Fix logic specifically for LTX-2.3 so it will be more automatic to find x2 or x1.5 resolutions.
* New history view on iPad with more log-like behavior.
Build 883646357v1.20260322.02026-03-23 21:11:29
* Fix FLUX.2 [klein] 9B KV cannot use batch size or text guidance != 1 correctly.
* Better Hires Fix logic specifically for LTX-2.3 so it will be more automatic to find x2 or x1.5 resolutions.
* New history view on iPad with more log-like behavior.
Build 883381853v1.20260314.02026-03-16 16:41:20
* Support LTX-2.3 for crisp clean audio and video generation;
* Support FLUX.2 [klein] 9B KV for up to 2x speed-up during image editing;
* Speed up video model's decoding performance.
Build 883039561v1.20260304.02026-03-05 17:57:34
* Fix metadata for LTX-2 models.
* Fix Z-Image LoRA trained by OneTrainer import issues.
* Revert to index to the last frame for generated videos to facilitate some existing scripts workflow.
* Fix Compute Units calculation issues for FLUX.1.
Build 883001110v1.20260303.02026-03-04 22:29:17
* Fixed a crash when generate LTX-2 video and enabled "Save Generated Media to";
* New Compute Units formulation that encourages video generation with Cloud Compute;
* New audio playback UI for video with audio.
Build 882880786v1.20260228.02026-03-03 23:46:25
* Support LTX-2 models for video generation with audio builtin, support both I2V and T2V mode, as well as importing their LoRAs;
* Fixed FLUX.2 issues related to loading LoRAs as side-path;
* Fixed FLUX.2 issues with tiled diffusion.
Build 882261841v1.20260207.02026-02-11 00:41:46
* Support Z Image Base series models;
* Support high quality preview on selected devices for Z Image, FLUX.1, HiDream, Qwen Image, FLUX.2;
* Fix importing some Z Image LoRAs.
Build 881682869v1.20260120.02026-01-22 21:49:27
* Support FLUX.2 [klein] series of models (4B, 9B, 4B Base, 9B Base);
* Support both import LoRAs and import models for FLUX.2 [klein] series.
Build 881171056v1.20260105.02026-01-07 19:37:33
* Support Qwen Image 2512 in FP16 and inside Official Models.
* Improve realism on Qwen Image models by fixing a precision problem.
Build 880965517v1.20251223.02025-12-28 15:57:56
* Added support for Qwen Image Edit 2511;
* Added support for Qwen Image Layered.
Build 880858424v1.20251219.02025-12-22 19:15:06
* Add support for FLUX.2, including model import and import LoRAs for FLUX.2 models.
Build 880634012v1.20251212.02025-12-15 18:32:21
* Improve compatibility with Z Image Turbo LoRAs and finetunes.
* Fix OS 26.2 specific issue with neural accelerators.
Build 880433786v1.20251207.02025-12-09 15:19:57
* Add support for Z Image Turbo, including model importing and LoRA support.
* Turn off automatic fallback to local rendering if remote rendering failed.
Build 880249693v1.20251201.02025-12-02 17:23:26
Hotfix a bug in 1.20251124.0 that stopping a remote generation would trigger a local generation.
Build 880054229v1.20251124.02025-12-01 01:03:03
* Support converting to 8-bit or quantized bit for models from the app.
* More error message on when Cloud Compute failed.
* Some minor UI fixes.
Build 879847710v1.20251117.12025-11-18 17:15:08
* Fix performance regressions on OS 26 for pre-A19 Pro era devices;
* Use Neural Accelerators for BF16 on A19 Pro / M5 devices;
* Fix cold start slowness for Neural Accelerators;
* LoRA import now respects your version selection more.
Build 879571740v1.20251107.12025-11-10 16:42:48
* Metal FlashAttention v2.5 w/ Neural Accelerators (preview) with breakthrough performance on M5 iPad / A19 Pro iPhones;
* Support import AuraFlow models;
* More compatibility improvement on importing Qwen Image LoRAs;
* Some minor UI adjustments.
Build 878935839v1.20251014.02025-10-15 16:03:07
* Fix SDXL training on M3 / M4;
* Support import Wan 2.2 5B models and LoRAs.
Price history by region
Currency
Show
Columns
Search
Availability 163 of 163 storefronts
| Region | Language | Price | Ratings | Avg | Version |
|---|---|---|---|---|---|
| US | en-US | Free | 721 | 4.5 | 1.20260518.2 |
| CN | zh-Hans-CN | Free | 417 | 4.6 | 1.20260518.2 |
| DE | de-DE | Free | 227 | 4.5 | 1.20260518.2 |
| JP | ja | Free | 131 | 4.4 | 1.20260518.2 |
| RU | ru | Free | 101 | 4.3 | 1.20260518.2 |
| GB | en-GB | Free | 86 | 4.3 | 1.20260518.2 |
| CA | en-CA | Free | 79 | 4.7 | 1.20260518.2 |
| TW | zh-Hant-TW | Free | 66 | 4.8 | 1.20260518.2 |
| FR | fr-FR | Free | 54 | 4.4 | 1.20260518.2 |
| ES | es-ES | Free | 53 | 4.9 | 1.20260518.2 |
| HK | zh-Hant-HK | Free | 51 | 4.8 | 1.20260518.2 |
| AU | en-AU | Free | 50 | 4.4 | 1.20260518.2 |
| TH | en-GB | Free | 49 | 4.0 | 1.20260518.2 |
| IN | en-GB | Free | 47 | 4.4 | 1.20260518.2 |
| IT | it | Free | 43 | 4.4 | 1.20260518.2 |
| KR | ko | Free | 39 | 4.8 | 1.20260518.2 |
| NL | nl | Free | 39 | 4.7 | 1.20260518.2 |
| CH | de-CH | Free | 25 | 4.9 | 1.20260518.2 |
| PL | en-GB | Free | 25 | 4.2 | 1.20260518.2 |
| UA | en-GB | Free | 24 | 4.6 | 1.20260518.2 |
| MX | es-MX | Free | 19 | 4.8 | 1.20260518.2 |
| TR | en-GB | Free | 18 | 4.1 | 1.20260518.2 |
| SE | sv | Free | 16 | 4.4 | 1.20260518.2 |
| BR | pt-BR | Free | 15 | 4.7 | 1.20260518.2 |
| AT | de-DE | Free | 14 | 5.0 | 1.20260518.2 |
| ID | en-GB | Free | 14 | 4.9 | 1.20260518.2 |
| SG | en-GB | Free | 12 | 4.6 | 1.20260518.2 |
| ZA | en-GB | Free | 12 | 3.9 | 1.20260518.2 |
| FI | en-GB | Free | 11 | 4.7 | 1.20260518.2 |
| MY | en-GB | Free | 11 | 4.7 | 1.20260518.2 |
| VN | en-GB | Free | 11 | 4.9 | 1.20260518.2 |
| CZ | en-GB | Free | 10 | 5.0 | 1.20260518.2 |
| PH | en-GB | Free | 10 | 4.4 | 1.20260518.2 |
| PT | pt-PT | Free | 9 | 4.8 | 1.20260518.2 |
| BE | en-GB | Free | 8 | 4.6 | 1.20260518.2 |
| BY | en-GB | Free | 7 | 3.3 | 1.20260518.2 |
| HU | en-GB | Free | 7 | 4.3 | 1.20260518.2 |
| KZ | en-GB | Free | 7 | 5.0 | 1.20260518.2 |
| NZ | en-AU | Free | 7 | 4.9 | 1.20260518.2 |
| RO | en-GB | Free | 7 | 5.0 | 1.20260518.2 |
| DK | en-GB | Free | 6 | 4.3 | 1.20260518.2 |
| NO | en-GB | Free | 5 | 5.0 | 1.20260518.2 |
| AE | en-GB | Free | 4 | 4.8 | 1.20260518.2 |
| EG | en-GB | Free | 3 | 3.7 | 1.20260518.2 |
| LU | en-GB | Free | 3 | 5.0 | 1.20260518.2 |
| EE | en-GB | Free | 2 | 5.0 | 1.20260518.2 |
| IQ | en-GB | Free | 2 | 3.0 | 1.20260518.2 |
| KG | en-GB | Free | 2 | 5.0 | 1.20260518.2 |
| LK | en-GB | Free | 2 | 4.0 | 1.20260518.2 |
| LT | en-GB | Free | 2 | 5.0 | 1.20260518.2 |
| MD | en-GB | Free | 2 | 5.0 | 1.20260518.2 |
| NP | en-GB | Free | 2 | 4.0 | 1.20260518.2 |
| PE | es-MX | Free | 2 | 5.0 | 1.20260518.2 |
| PK | en-GB | Free | 2 | 5.0 | 1.20260518.2 |
| RS | en-GB | Free | 2 | 5.0 | 1.20260518.2 |
| SI | en-GB | Free | 2 | 5.0 | 1.20260518.2 |
| AZ | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| BG | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| BN | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| CR | es-MX | Free | 1 | 5.0 | 1.20260518.2 |
| DO | es-MX | Free | 1 | 5.0 | 1.20260518.2 |
| GE | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| GT | es-MX | Free | 1 | 5.0 | 1.20260518.2 |
| HR | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| IE | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| IS | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| KW | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| LV | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| MA | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| MK | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| MT | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| PA | es-MX | Free | 1 | 5.0 | 1.20260518.2 |
| QA | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| SK | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| SV | es-MX | Free | 1 | 5.0 | 1.20260518.2 |
| TT | en-GB | Free | 1 | 5.0 | 1.20260518.2 |
| AF | en-GB | Free | 0 | — | 1.20260518.2 |
| AG | en-GB | Free | 0 | — | 1.20260518.2 |
| AI | en-GB | Free | 0 | — | 1.20260518.2 |
| AL | en-GB | Free | 0 | — | 1.20260518.2 |
| AM | en-GB | Free | 0 | — | 1.20260518.2 |
| AO | en-GB | Free | 0 | — | 1.20260518.2 |
| BA | en-GB | Free | 0 | — | 1.20260518.2 |
| BB | en-GB | Free | 0 | — | 1.20260518.2 |
| BF | en-GB | Free | 0 | — | 1.20260518.2 |
| BJ | en-GB | Free | 0 | — | 1.20260518.2 |
| BM | en-GB | Free | 0 | — | 1.20260518.2 |
| BO | es-MX | Free | 0 | — | 1.20260518.2 |
| BS | en-GB | Free | 0 | — | 1.20260518.2 |
| BT | en-GB | Free | 0 | — | 1.20260518.2 |
| BW | en-GB | Free | 0 | — | 1.20260518.2 |
| BZ | en-GB | Free | 0 | — | 1.20260518.2 |
| CD | en-GB | Free | 0 | — | 1.20260518.2 |
| CG | en-GB | Free | 0 | — | 1.20260518.2 |
| CI | fr-FR | Free | 0 | — | 1.20260518.2 |
| CM | fr-FR | Free | 0 | — | 1.20260518.2 |
| CV | en-GB | Free | 0 | — | 1.20260518.2 |
| DM | en-GB | Free | 0 | — | 1.20260518.2 |
| DZ | en-GB | Free | 0 | — | 1.20260518.2 |
| FJ | en-GB | Free | 0 | — | 1.20260518.2 |
| FM | en-GB | Free | 0 | — | 1.20260518.2 |
| GA | fr-FR | Free | 0 | — | 1.20260518.2 |
| GD | en-GB | Free | 0 | — | 1.20260518.2 |
| GH | en-GB | Free | 0 | — | 1.20260518.2 |
| GM | en-GB | Free | 0 | — | 1.20260518.2 |
| GR | en-GB | Free | 0 | — | 1.20260518.2 |
| GW | en-GB | Free | 0 | — | 1.20260518.2 |
| GY | en-GB | Free | 0 | — | 1.20260518.2 |
| HN | es-MX | Free | 0 | — | 1.20260518.2 |
| JM | en-GB | Free | 0 | — | 1.20260518.2 |
| KE | en-GB | Free | 0 | — | 1.20260518.2 |
| KH | en-GB | Free | 0 | — | 1.20260518.2 |
| KN | en-GB | Free | 0 | — | 1.20260518.2 |
| KY | en-GB | Free | 0 | — | 1.20260518.2 |
| LA | en-GB | Free | 0 | — | 1.20260518.2 |
| LC | en-GB | Free | 0 | — | 1.20260518.2 |
| LR | en-GB | Free | 0 | — | 1.20260518.2 |
| LY | en-GB | Free | 0 | — | 1.20260518.2 |
| ME | en-GB | Free | 0 | — | 1.20260518.2 |
| MG | en-GB | Free | 0 | — | 1.20260518.2 |
| ML | en-GB | Free | 0 | — | 1.20260518.2 |
| MM | en-GB | Free | 0 | — | 1.20260518.2 |
| MN | en-GB | Free | 0 | — | 1.20260518.2 |
| MO | zh-Hant-HK | Free | 0 | — | 1.20260518.2 |
| MR | en-GB | Free | 0 | — | 1.20260518.2 |
| MS | en-GB | Free | 0 | — | 1.20260518.2 |
| MU | en-GB | Free | 0 | — | 1.20260518.2 |
| MV | en-GB | Free | 0 | — | 1.20260518.2 |
| MZ | en-GB | Free | 0 | — | 1.20260518.2 |
| NA | en-GB | Free | 0 | — | 1.20260518.2 |
| NE | en-GB | Free | 0 | — | 1.20260518.2 |
| NG | en-GB | Free | 0 | — | 1.20260518.2 |
| NI | es-MX | Free | 0 | — | 1.20260518.2 |
| NR | en-GB | Free | 0 | — | 1.20260518.2 |
| OM | en-GB | Free | 0 | — | 1.20260518.2 |
| PG | en-GB | Free | 0 | — | 1.20260518.2 |
| PW | en-GB | Free | 0 | — | 1.20260518.2 |
| PY | es-MX | Free | 0 | — | 1.20260518.2 |
| RW | en-GB | Free | 0 | — | 1.20260518.2 |
| SB | en-GB | Free | 0 | — | 1.20260518.2 |
| SC | en-GB | Free | 0 | — | 1.20260518.2 |
| SL | en-GB | Free | 0 | — | 1.20260518.2 |
| SN | en-GB | Free | 0 | — | 1.20260518.2 |
| SR | en-GB | Free | 0 | — | 1.20260518.2 |
| ST | en-GB | Free | 0 | — | 1.20260518.2 |
| SZ | en-GB | Free | 0 | — | 1.20260518.2 |
| TC | en-GB | Free | 0 | — | 1.20260518.2 |
| TD | en-GB | Free | 0 | — | 1.20260518.2 |
| TJ | en-GB | Free | 0 | — | 1.20260518.2 |
| TM | en-GB | Free | 0 | — | 1.20260518.2 |
| TN | en-GB | Free | 0 | — | 1.20260518.2 |
| TO | en-GB | Free | 0 | — | 1.20260518.2 |
| TZ | en-GB | Free | 0 | — | 1.20260518.2 |
| UG | en-GB | Free | 0 | — | 1.20260518.2 |
| UZ | en-GB | Free | 0 | — | 1.20260518.2 |
| VC | en-GB | Free | 0 | — | 1.20260518.2 |
| VE | es-MX | Free | 0 | — | 1.20260518.2 |
| VG | en-GB | Free | 0 | — | 1.20260518.2 |
| VU | en-GB | Free | 0 | — | 1.20260518.2 |
| XK | en-GB | Free | 0 | — | 1.20260518.2 |
| YE | en-GB | Free | 0 | — | 1.20260518.2 |
| ZM | en-GB | Free | 0 | — | 1.20260518.2 |
| ZW | en-GB | Free | 0 | — | 1.20260518.2 |
Change log 29 changes · PY
✎
releaseNotes updated
* Add interrogator model "Qwen 3.5 4B".
* Add dedicated mode selector "Auto" / "Generation" / "Edit".
* Support "Sign in with Google".
* Some other small visual
✎
description updated
Unleash your imagination with Draw Things, the powerful offline AI creation tool for your iPhone or iPad. Create stunning images privately, right on your device
✎
subtitle updated
Your Private AI Art Studio
✎
name updated
Draw Things: Offline AI Art
◆
Version 1.20260518.2 released
* Add interrogator model "Qwen 3.5 4B".
* Add dedicated mode selector "Auto" / "Generation" / "Edit".
* Support "Sign in with Google".
* Some other small visual updates.
◆
Version 1.20260501.0 released
* Hotfix for new segmented downloader issue.
◆
Version 1.20260430.0 released
* Support SeedVR2 as one of the diffusion model for 1-step upscale / restoration work.
* Fix Anima LLM adapter related issues.
* Switch to better newer causal / mask attention kernels.
* Support export LoRAs for Qwen, Z Image, FLUX.2 and ER
◆
Version 1.20260418.1 released
* Added support for ERNIE Image, including LoRA import, training and model import.
* Added support for Anima Preview 3, including LoRA import, training and model import.
* Updated LTX-2.3 [distilled] to LITX-2.3 [distilled] 1.1.
* Fixed cra
◆
Version 1.20260410.1 released
* Add 8-bit S support for M4 / A18 Pro devices, reaching up to 2x speed-up for these device.
* Support LoRA training for FLUX.2 [klein] 4B / 9B / Z Image / Qwen Image as base model.
* Fix audio sync issue during in-app playback.
* Fix some
◆
Version 1.20260330.0 released
* Added a new quantization variant: 8-bit S, which on A19 Pro / M5 is about 1.5x to 2x faster than Exact or 8-bit variant.
* Fixed issue when importing models, it was no longer respecting External Folder selection.
* Support Bridge Mode to
◆
Version 1.20260323.0 released
* Support export video as HEVC;
* Fix new History item menu issue.
◆
Version 1.20260322.0 released
* Fix FLUX.2 [klein] 9B KV cannot use batch size or text guidance != 1 correctly.
* Better Hires Fix logic specifically for LTX-2.3 so it will be more automatic to find x2 or x1.5 resolutions.
* New history view on iPad with more log-like b
◆
Version 1.20260320.1 released
* Fix FLUX.2 [klein] 9B KV cannot use batch size or text guidance != 1 correctly.
* Better Hires Fix logic specifically for LTX-2.3 so it will be more automatic to find x2 or x1.5 resolutions.
* New history view on iPad with more log-like b
◆
Version 1.20260314.0 released
* Support LTX-2.3 for crisp clean audio and video generation;
* Support FLUX.2 [klein] 9B KV for up to 2x speed-up during image editing;
* Speed up video model's decoding performance.
◆
Version 1.20260304.0 released
* Fix metadata for LTX-2 models.
* Fix Z-Image LoRA trained by OneTrainer import issues.
* Revert to index to the last frame for generated videos to facilitate some existing scripts workflow.
* Fix Compute Units calculation issues for FLUX.
◆
Version 1.20260303.0 released
* Fixed a crash when generate LTX-2 video and enabled "Save Generated Media to";
* New Compute Units formulation that encourages video generation with Cloud Compute;
* New audio playback UI for video with audio.
◆
Version 1.20260228.0 released
* Support LTX-2 models for video generation with audio builtin, support both I2V and T2V mode, as well as importing their LoRAs;
* Fixed FLUX.2 issues related to loading LoRAs as side-path;
* Fixed FLUX.2 issues with tiled diffusion.
◆
Version 1.20260207.0 released
* Support Z Image Base series models;
* Support high quality preview on selected devices for Z Image, FLUX.1, HiDream, Qwen Image, FLUX.2;
* Fix importing some Z Image LoRAs.
◆
Version 1.20260120.0 released
* Support FLUX.2 [klein] series of models (4B, 9B, 4B Base, 9B Base);
* Support both import LoRAs and import models for FLUX.2 [klein] series.
◆
Version 1.20260105.0 released
* Support Qwen Image 2512 in FP16 and inside Official Models.
* Improve realism on Qwen Image models by fixing a precision problem.
◆
Version 1.20251223.0 released
* Added support for Qwen Image Edit 2511;
* Added support for Qwen Image Layered.
◆
Version 1.20251219.0 released
* Add support for FLUX.2, including model import and import LoRAs for FLUX.2 models.
◆
Version 1.20251212.0 released
* Improve compatibility with Z Image Turbo LoRAs and finetunes.
* Fix OS 26.2 specific issue with neural accelerators.
◆
Version 1.20251207.0 released
* Add support for Z Image Turbo, including model importing and LoRA support.
* Turn off automatic fallback to local rendering if remote rendering failed.
◆
Version 1.20251201.0 released
Hotfix a bug in 1.20251124.0 that stopping a remote generation would trigger a local generation.
◆
Version 1.20251124.0 released
* Support converting to 8-bit or quantized bit for models from the app.
* More error message on when Cloud Compute failed.
* Some minor UI fixes.
◆
Version 1.20251117.1 released
* Fix performance regressions on OS 26 for pre-A19 Pro era devices;
* Use Neural Accelerators for BF16 on A19 Pro / M5 devices;
* Fix cold start slowness for Neural Accelerators;
* LoRA import now respects your version selection more.
◆
Version 1.20251107.1 released
* Metal FlashAttention v2.5 w/ Neural Accelerators (preview) with breakthrough performance on M5 iPad / A19 Pro iPhones;
* Support import AuraFlow models;
* More compatibility improvement on importing Qwen Image LoRAs;
* Some minor UI adjus
◆
Version 1.20251014.0 released
* Fix SDXL training on M3 / M4;
* Support import Wan 2.2 5B models and LoRAs.