Strata v0.1.11

Chat settings for all your apps.

• "Use for other apps too" (new). In the browser app, Chat → gear icon (Sampling) has a new switch at the bottom. Turn it on and press Apply, and coding agents like omp and other API clients get your Chat settings - thinking level, temperature, top-p, top-k, max tokens, seed and the experimental speed projection - for anything they don't set themselves, from their next request (no restart needed). What an app sends itself still wins. Turn the switch off to let apps use their own settings again. The settings are kept on the server between starts, and only Strata's own page can change them (and only with the API key, when you set one).
• Fixed: in the Sampling drawer, the experimental speed projection switch was squeezed by its longer label and cut off at the edge.

Updating: get the latest files (git pull or download the repo again) and start Strata as usual. The engine is unchanged since 0.1.10, so nothing is downloaded again.

The ready-made Strata engine for Windows (RTX 30 / 40 / 50: sm_86, sm_89, sm_120 + PTX), CUDA 13.0.

You don't need to download this yourself: START-HERE.bat fetches it (and NVIDIA's cuBLAS from pip), so no compiler or CUDA Toolkit is needed, only an NVIDIA driver 580 or newer.

Contents: strata.exe (the engine), strata-vision.exe (the optional image encoder), BUILD.json.