DiffusionBear: MLX-powered macOS app runs FLUX.2 & Krea 2 on 16GB Macs
Puzzleheaded_Note739 · reddit · 2026-10-12
A developer released DiffusionBear v0.3.8 beta (MIT-licensed), a native macOS image generation studio built on Apple's MLX and optimized for M-series chips. Its 4-bit quantization keeps FLUX.2-Klein (4B/9B), Krea 2 Turbo 13B, Z-Image Turbo 6B, Qwen-Image-2.1-Turbo and SDXL resident in 16GB unified memory without SSD swapping. It supports drag-and-drop .safetensors LoRAs (up to 16, Civitai sync), runs standalone on Metal with no Python env, and writes full metadata (engine, seed, steps, quantization, time) into PNG sidecars. Beta testers wanted for benchmarks, LoRA compat and Metal bugs.
More from Infra
- OpenRouter token traffic explodes from 2T to 379T/month, open-weight models at 75% — Beth_Kindig · 2026-10-12
- ASUS RTX 5090 listed at €11,079 in Germany, only 3 units left in stock — janusch_patas · 2026-10-12
- Modular Claims Mojo Kernels Beat FlashAttention 4, Launches MAX Inference Framework — AI Engineer · 2026-10-12
- DeepSeek-V4.1-Flash on 2x DGX Spark: TP2 Patches Cut First Token From 33.8s to 4.9s — rez0__ · 2026-10-12
- Ed Zitron: AI needs $425B annual revenue to justify hyperscaler capex, $260B short — SumitGup · 2026-10-12
- mitsuhiko: No one-size-fits-all sandbox for agents — VMs, sandboxes, durable objects each have their place — mitsuhiko · 2026-10-12