HexAnima: Mage-Flow-Edit-Turbo pack

Microsoft's Mage-Flow-Edit-Turbo (4B, 4 steps) with its Qwen3-VL-4B text encoder, vision tower and Mage-VAE, converted for the Hexagon NPU of Snapdragon 8 Gen 1 / 8+ Gen 1 phones. It does text-to-image and photo editing in about 10-12 s per picture. The app (https://github.com/ayanoby/hexanima) downloads these files one by one with resume and sha256 verification (manifest.json). They are not usable with other tools.

About 8.2 GB: text encoder 4.0 GB, DiT 3.1 GB, vision tower 0.7 GB, VAE 0.3 GB, token embeddings and glue.

MIT license (LICENSE); the Qwen3-VL parts are Apache-2.0. See NOTICE.md for authors, sources and changes. HexAnima is an independent, unofficial project, not affiliated with the authors named in NOTICE.md.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for ayanoby/hexanima-mage-packs

Finetuned
(2)
this model