DiffusionBear v0.5.0 released: pure MLX runtime, size halved
Start here
DiffusionBear released v0.5.0, changing the runtime to pure MLX and removing PyTorch, reducing app size from 1.3 GB to 615 MB, adding virtual box, Kromoss 5B support, RGBA transparency, and Agentic REST API; the publisher says total time for 220 benchmark tests decreased by 58.5%. This version only supports Apple Silicon (M1/M2/M3/M4/Ultra), requires macOS 26.2 or higher, recommends 16 GB unified memory; model weights are not bundled with the app, downloaded from Hugging Face or Civitai on first use.
Written by AI from the reports · updated 3 days ago
Report timeline
Follow the reports to see the event from other sides.
- Reddit StableDiffusion · 创作者与工作流分享DiffusionBear v0.5.0 released: pure MLX runtime, virtual box, and Agentic REST API
DiffusionBear v0.5.0 released, runtime changed to pure MLX and PyTorch removed, app size reduced from 1.3 GB to 615 MB, added virtual box, Kromoss 5B support, RGBA transparency, and Agentic REST API. Memory lifecycle changed to 3/4-phase sequential execution, total time for 220 benchmark tests decreased by 58.5%.
Heat trend for this event
- Current
- 3
- Peak
- 10Nov 16 08:00
- 24h change
- -51%
The trend compares only sources observed the whole time, so the range may be smaller than the current heat figure. Hover or tap the chart for each hour. Left and right arrow keys move through it.