Autobot 1.6.2

On-device image generation fixed for Local Dream NPU models, sturdier downloads and engines, safer agents, pixel-exact dots, and much less text on iPhone.

Fixed: image generation on Snapdragon (Android)

• NPU image models now run. A phone whose only image models were Local Dream / xororz NPU packages (SD 1.5, SDXL, Illustrious and others) never got a "This phone" image backend, so none of them could be used. They now appear in Imagine and Studio.
• Builds picked like Local Dream does. Snapdragon 8 Elite Gen 5 now gets the 8gen2 builds (SDXL: 8gen3), matching Local Dream. If you installed an NPU upscaler before, reinstall it so you get the right build.
• SDXL uses less memory. Only the model in use stays loaded. SDXL frees its text encoders before loading the UNet, and the VAE encoder only loads for img2img and inpainting.
• Clear errors. A failure now names the stage (loading, text encoder, UNet) and quotes the engine's own message. Models › Compute › Report includes a live NPU check: how far the NPU setup got, the runtime version, the chip's library, and a test load of your smallest NPU model. If something still fails, copy that report and send it.

Downloads and engines (Android)

• A cancelled download no longer comes back after the app is killed. Pause and cancel take effect at once, even on a stalled connection.
• Resuming a file that changed on the server restarts it cleanly instead of splicing two versions together.
• A disk that fills up pauses with a storage message. Gated Hugging Face models tell you to sign in and accept the terms.
• Models whose files were deleted outside the app are detected and offered for re-download (only the missing files).
• A crashed engine is released and restarted cleanly. Cancelling during model load now works. Broken or wrong model files are named before loading.
• Tapping a Whisper model opens voice mode.

Agents (both platforms)

• Every tool that can reach the network is checked for private data, and an approval prompt can no longer override a refusal.
• Approvals cancelled midway are logged correctly. Replies the model cut short are no longer retried as empty.
• Privacy receipts list every model your data went to, including subagents and summaries.
• Studio remote runs time out and cancel properly.

Look and feel

• Pixel-exact dots (both platforms). Bars, loaders, living cards and backgrounds sit on a whole-pixel grid: evenly spaced, centred, and never clipped. Tests check every dot to within half a pixel at several screen densities.
• Less text (iPhone). Settings rows show status instead of subtitles, with explanations behind info buttons, grouped into six sections. Models, Run, Imagine, Privacy, Studio and more are trimmed. Generating with a downloaded model takes one tap fewer.
• Android: clear states when model search is empty (offline, error, no results), with a shortcut to the network setting. Fixed "1 SESSIONS" and the double model size. Status-bar colours in Studio corrected. Setup problems each get one clear message and a fix button.

Faster

• Opening a long chat on Android no longer decodes its history on the main thread.
• Studio thumbnails use at most 24 MB of memory.

Files

• autobot-1.6.2-release.apk: signed release (ai.wckd.autobot, arm64-v8a, Android 14+). Signer certificate SHA-256 9a1378acf9587885e7920355558c85a1ec0e89304f58c153feebf90fdee7f3f1.
• autobot-1.6.2-debug.apk
• autobot-1.6.2-mapping.txt: R8 mapping, for reading crash reports
• SHA256SUMS.txt

iOS: TestFlight build 1.6.2.