Mouserat/Anything-V5-mnn

Anything V5 (MNN)

This repository provides an MNN-format conversion of stablediffusionapi/anything-v5 for use with the nezumi-ai image generation engine — a fully offline AI chat app for Android supporting on-device image generation.

A Windows CLI (nezumi-ai-sd-cli) is also provided for testing/debugging on desktop, but the primary target platform is Android. This model is not bundled with the app — users download it separately and individually agree to its license terms.

Platform note: nezumi-ai-sd-cli currently builds for Windows only (.exe). A Linux build is planned.

Variants

SD1.5 models are distributed in a single quantization variant (unlike the SDXL/Illustrious line, which offers int4/int8 variants):

File	UNet quantization	Size
Anything-V5-mnn-int8-block32.zip	8-bit, block size 32	~1.23 GB

Model Provenance

Field	Value
Base model	stablediffusionapi/anything-v5
Original source	Popular anime-style SD1.5 checkpoint originally distributed on Civitai; this HF repo is a diffusers-format re-conversion published by ModelsLab / stablediffusionapi.com for API use, not a first-party upload by the original checkpoint author.
Format	MNN (clip_v2.mnn + .weight, unet.mnn + .weight, vae_decoder_fp16.mnn + .weight, token_emb.bin, pos_emb.bin, tokenizer.json)
Conversion tool	convert_hf_to_mnn_sd.py (nezumi-ai)

Provenance note: unlike other models in this collection, we could not confirm a first-party Hugging Face repository from the original Anything V5 creator. The stablediffusionapi/anything-v5 repo used here is a third-party diffusers-format re-packaging (11GB, converted by ModelsLab for their paid API product) rather than an upload from the original author. The declared creativeml-openrail-m license tag is taken at face value from that repo. If you know of an authoritative first-party source for Anything V5, please open a discussion on this repo.

Conversion settings

python convert_hf_to_mnn_sd.py \
    --model stablediffusionapi/anything-v5 \
    --out ./out/anything-v5 \
    --size 512 \
    --unet-bits 8 \
    --unet-block 32 \
    --clip-bits 8 \
    --vae-bits 8

No fine-tuning or retraining was performed — weights are unchanged from the original checkpoint aside from format conversion and the quantization above.

Output files

clip_v2.mnn                0.13 MB
clip_v2.mnn.weight       148.92 MB
model.json                 0.00 MB
pos_emb.bin                 0.23 MB
token_emb.bin              72.38 MB
tokenizer.json               2.12 MB
unet.mnn                    1.13 MB
unet.mnn.weight          911.38 MB
vae_decoder_fp16.mnn        0.22 MB
vae_decoder_fp16.mnn.weight 94.38 MB
TOTAL                    1230.89 MB

License

• Declared license: CreativeML Open RAIL-M (full text), as tagged on the stablediffusionapi/anything-v5 repository.
• Redistribution: Permitted under the declared license.
• Commercial use: Permitted within the terms of the RAIL-M license.
• Attribution: Required — credit to the original Anything V5 author(s) and to stablediffusionapi.com / ModelsLab for the diffusers-format conversion this repo builds on.

Please read and comply with the original license before using this model. Because the upstream repository used here is a third-party re-packaging rather than a confirmed first-party source (see provenance note above), exercise extra care if using this model commercially — verify the license and origin independently if it matters for your use case.

This checkpoint inherits the declared RAIL-M use-based restrictions in full (see Attachment A of the full license text), including prohibitions on use for exploiting minors, generating disinformation, harassment, discrimination, unauthorized medical advice, and law-enforcement/immigration profiling.

Note: the conversion script itself is part of the nezumi-ai project and licensed separately under LGPL v3 / a commercial license (see LICENSE.md). That license applies to the code, not to this model checkpoint.

Requirements (Android)

Item	Minimum	Recommended
Android Version	12 (API 31)	14+ (API 34+)
RAM	4GB	6GB+
Storage	3GB free	5GB+
GPU	Optional — OpenCL-capable GPU (Adreno, Mali, PowerVR)	Recommended

SD1.5 requirements are lower than SDXL/Illustrious (8GB RAM minimum) — see that model's README for comparison.

Usage

Android (primary)

Used automatically by the nezumi-ai app's image-generation feature (MNN backend, GPU/OpenCL → CPU fallback). Download/select this model from within the app; manual extraction is not required on Android.

Windows CLI (testing/debugging)

Distributed as a zip archive. Extract it before use — in PowerShell:

Expand-Archive Anything-V5-mnn-int8-block32.zip C:\sd-model

Then run:

nezumi-ai-sd-cli "C:\sd-model" "1girl, masterpiece, best quality, anime style, detailed" --steps 20 --width 512 --height 768 --backend cpu --out out.png

Options

Option	Description	Default
<model_path>	Path to the extracted MNN model folder	—
<prompt>	Text prompt	—
--negative <text>	Negative prompt	empty
--width <n> / --height <n>	Image size	512 / 512
--steps <n>	Sampling steps	20
--cfg <f>	CFG scale	7.0
--seed <n>	Seed (negative = random)	-1
--scheduler <name>	Sampling scheduler: euler|ddim|dpm|dpm++2m|dpm++2m-karras|lcm|eulera|unipc	dpm++2m
--backend <name>	cpu|opencl	cpu
--out <path>	Output path. .ppm always works; .png requires stb_image_write.h	—

Danbooru-style tag prompts work well with this model. eulera or dpm++2m-karras at CFG 7–8 with 20–30 steps is a reasonable starting point.

Roadmap

• Linux build of nezumi-ai-sd-cli
• macOS build
• SDXL support
• Additional quantization variants

Disclaimer

This is an unofficial, community conversion and is not affiliated with or endorsed by the original Anything V5 creator(s) or by ModelsLab / stablediffusionapi.com.