Image-Text-to-Text
Transformers.js
ONNX
lfm2_vl
ad-detection
ad-blocking
webgpu
axera
npu
edge
transformers-js
lfm2.5
vision-language-model
browser
conversational
Instructions to use TheGarageDev/Minus-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use TheGarageDev/Minus-v0.1 with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('image-text-to-text', 'TheGarageDev/Minus-v0.1');
Commit History
2 days ago
-
Add the Axera AX650N/AX8850 NPU build (axera/) + model card section
Verified signatureSame iteration-28 weights as onnx/, compiled with Pulsar2 (U16, FP32 RMSNorm chain, NPU3). This is what the minus HDMI device runs on a Radxa AICore AX-M1: ~360 ms/frame, 95.6% on an 800-image held-out set.42538b4
-
model card: add Buy Me a Coffee support link
Verified signature0db3dfc -
model card: remove demo-space link
Verified signature43dbe82 -
model card: link the WebGPU demo Space
Verified signatured34b224 -
model card: trim streaming-data blurb
Verified signaturefb79f45 -
model card: add screenshots
Verified signature99dd319 -
assets: review page screenshot
Verified signature6169abb -
assets: real-site screenshot
Verified signature3a429e4 -
Super-squash branch 'main' using huggingface_hub
f622778