SHIPPED · v0.11.0
1LocalAITool v0.11.0 — the graphics card used and accounted for, your own picture checkpoints, and StoicSoft bundle keys
New
Pictures and 3D use an NVIDIA card on Windows. Pictures fetch the CUDA build with its runtime, then Vulkan, then the processor, so an RTX card no longer draws on the processor. 3D installs PyTorch built for the card. A driver older than 525 skips CUDA rather than failing on the first picture.
Text uses an NVIDIA card on Linux. The engine now fetches the Vulkan build, which every NVIDIA and AMD driver runs. An engine installed before is replaced the next time the app starts, before anything can start the old one.
macOS 15 makes pictures and 3D models again, without being told to upgrade to macOS 26. The newest picture build made for macOS 15 is found further back in the release list. A known-good build is kept as a last resort, so this keeps working as newer releases push it further back, and when GitHub limits how often the list can be read.
Running on the processor is never silent. When there is a card and text, pictures or 3D end up on the processor, a dialog says so, with the reason in the engine's own words, *Logs and a report*, and *Try the card again*. The sidebar light is green on the card and amber off it. This also covers a text engine that starts but finds no card it can use — a missing Vulkan driver used to leave it on the processor with the light still green.
Try the card again really tries again. It restarts text on the card, fetches the card's picture build afresh with the next picture, and reinstalls the card's PyTorch for 3D when that was the cause. A fallback is no longer kept for good.
Graphics card only. A new switch under *Logs and crash reports*, the opposite of *Use the processor only*: when the card cannot do something, text, pictures and 3D stop and say why instead of moving to the processor.
This computer says what to expect before anything runs. One line per feature — text, pictures and video, 3D — says whether it runs on the card or the processor on this machine, and why.
Another program on the app's port is named. A banner says whether it is another copy of the app or some other program, instead of retrying in silence.
The speed test ends. It is capped at 96 tokens and three minutes, and reads the engine's own speed figure. It could sit on *Measuring this computer* for as long as the model kept writing.
A checkpoint you downloaded yourself can be picked wherever a picture is made. *Make a picture*, *Make a video* and *Make a 3D model* on Tasks gain *Use a checkpoint you downloaded yourself* under the model list: paste the link behind a Hugging Face file's download button, or add a .safetensors or diffusion .gguf already on this computer, and it is picked for that picture the moment it lands. The Pictures section on Models has the same controls, so a checkpoint no longer has to go in through the text models' *Add your own*.
Look for them finds picture apps' checkpoints. The scan now looks where ComfyUI (and its desktop app), Automatic1111, Forge, SD.Next, Fooocus, InvokeAI and Stability Matrix keep theirs, besides LM Studio, the Hugging Face cache and Downloads — and says which files are text models and which are checkpoints, so each kind is offered in its own place. A LoRA, a VAE, a text encoder or a ControlNet on its own is left out instead of listing as a checkpoint the worker cannot draw with.
Adding a checkpoint from another app's folder leaves that app its file. On the same drive the checkpoint is stored once and shared, so it takes no extra room and ComfyUI still finds it where it was. Across drives the file moves, as before.
Picture models in .gguf form are no longer taken for text models. Qwen-Image, AuraFlow and HunyuanVideo files from some converters listed under text; they now file under pictures.
Bundle keys work here. A key from the StoicSoft store — the kind that starts with SS- — turns on Pro the same way a Lemon Squeezy key does. Keys you already have keep working exactly as before.
The License page names your bundle. With a bundle key it says which one, for example *This copy of 1LocalAITool is Pro, through your StoicSoft Complete key*.
The other StoicSoft apps, in one card. Below the license, a card shows the bundles that include 1LocalAITool and what they save. With Pro it shows the discount on your next StoicSoft app instead; with a bundle, the other apps your key already covers and, when there is room, more devices. It folds down to a single line, or away.
Validating picks up a changed key. *Validate* now refreshes the plan name and the device limit, so a bundle or extra devices added to your key show up without activating again.