Launches
Anima Turbo gets an INT8 build for faster local anime images
Community reports 2 sources
A
Anima Turbo v1.1 Nunchaku
Model Oct 13, 2026
A community INT8 quantization of Anima Turbo v1.1 for local anime and illustration generation with Nunchaku.
Visit primary link
2 sources attached
What happened lately
Anima Turbo v1.1 Nunchaku timeline
Tonera’s Anima Turbo v1.1 Nunchaku repository provides an INT8 weight-and-activation quantized transformer, companion components, instructions and example images. It derives from CircleStone Labs and Comfy Org’s Anima, a text-to-image model aimed at anime and illustration. The author reports speed and memory improvements against the BF16 pipeline on three NVIDIA GPUs; these are project benchmarks, not an independent test. The weights inherit upstream non-commercial terms, and the documented installation currently uses a specific Linux, CUDA and Python build.