Rename package gelato_local -> quadembed, drop GELATO from project naming (credit stays in text)

README.md

@@ -13,13 +13,19 @@ tags:
 library_name: pytorch
 ---
 
-# gelato-local-nano
+# QuadEmbed
 
-A local, from-scratch-trained reproduction of **GELATO** (Geometry-preserving
-Embeddings via Locked Aligned TOwers), the architecture behind
-[jina-embeddings-v5-omni](https://arxiv.org/abs/2605.08384). It maps text,
-images, audio, and video into one shared 768-dimensional embedding space,
-trained end-to-end on a single 8GB consumer GPU (RTX 4060 laptop).
+A local, from-scratch-trained multimodal embedding model. "Quad" for the
+four modalities it covers: text, image, audio, and video, all mapped into
+one shared 768-dimensional embedding space, trained end-to-end on a single
+8GB consumer GPU (RTX 4060 laptop).
+
+The architecture reproduces **GELATO** (Geometry-preserving Embeddings via
+Locked Aligned TOwers), described by Jina AI in the paper behind
+[jina-embeddings-v5-omni](https://arxiv.org/abs/2605.08384). Full credit for
+the architecture and training recipe goes to that paper; this repo is an
+independent, from-scratch reproduction of it, not a copy of Jina's released
+weights.
 
 This is not a copy of Jina's released checkpoint — the three encoders below
 are frozen, publicly available source models, and only two small projector
@@ -70,12 +76,11 @@ stayed under 2.7GB — well inside an 8GB budget.
 
 These numbers come after seven rounds of iterating on the vision projector
 specifically (architecture fixes, data scale, data diversity, projector
-capacity) — full write-up of what did and didn't move the needle is in this
-project's [feasibility write-up and README](https://github.com/) *(local
-project — see the blog post that accompanies this model for the full
-round-by-round story)*. Short version: architecture wasn't the bottleneck
-after round 3; raw training data volume was — R@1 only moved once the
-image-caption training set was scaled to ~172k pairs (round 7).
+capacity). Short version: architecture wasn't the bottleneck after round 3;
+raw training data volume was — R@1 only moved once the image-caption
+training set was scaled to ~172k pairs (round 7). The full round-by-round
+write-up, with what did and didn't move the needle, lives in the source
+repo's README and feasibility notes (linked once published).
 
 ## Usage
 
@@ -84,10 +89,10 @@ pip install -r requirements.txt
 ```
 
 ```python
-from gelato_local import GelatoLocal
+from quadembed import QuadEmbed
 from PIL import Image
 
-model = GelatoLocal.from_pretrained("checkpoints")  # auto-detects cuda/cpu
+model = QuadEmbed.from_pretrained("checkpoints")  # auto-detects cuda/cpu
 
 text_embeds = model.embed_text(["a dog running on the beach"])
 image_embeds = model.embed_image([Image.open("photo.jpg").convert("RGB")])
@@ -137,7 +142,8 @@ Jina AI directly.
 
 ## Citation
 
-This repo reproduces the architecture described in:
+QuadEmbed reproduces the architecture described in Jina AI's GELATO paper.
+All credit for the original architecture and training recipe belongs there:
 
 ```
 @article{gelato2026,
@@ -146,5 +152,6 @@ This repo reproduces the architecture described in:
 }
 ```
 
-Not affiliated with or endorsed by Jina AI — an independent, from-scratch
-reproduction built for a learning/portfolio project.
+QuadEmbed is not affiliated with or endorsed by Jina AI — an independent,
+from-scratch reproduction of their published architecture, built as a
+learning and portfolio project.

examples/inference_example.py

@@ -8,9 +8,9 @@ Run from the repo root:
 """
 from PIL import Image
 
-from gelato_local import GelatoLocal
+from quadembed import QuadEmbed
 
-model = GelatoLocal.from_pretrained("checkpoints")  # device auto-detects cuda/cpu
+model = QuadEmbed.from_pretrained("checkpoints")  # device auto-detects cuda/cpu
 
 texts = ["a photo of a dog running on the beach", "a bowl of ramen noodles"]
 text_embeds = model.embed_text(texts)

{gelato_local → quadembed}/__init__.py

@@ -1,11 +1,12 @@
-"""gelato-local-nano: a local, from-scratch-trained reproduction of GELATO
-(the architecture behind jina-embeddings-v5-omni) -- frozen text/vision/audio
-encoders plus two small trained projectors, sharing one 768-dim embedding
-space across text, image, audio, and video.
+"""QuadEmbed: a local, from-scratch-trained multimodal embedding model
+covering all four modalities, text, image, audio, and video, in one shared
+768-dim space. Reproduces the architecture behind Jina AI's GELATO
+(jina-embeddings-v5-omni): frozen text/vision/audio encoders plus two small
+trained projectors.
 
-    from gelato_local import GelatoLocal
+    from quadembed import QuadEmbed
 
-    model = GelatoLocal.from_pretrained("checkpoints", device="cuda")
+    model = QuadEmbed.from_pretrained("checkpoints", device="cuda")
     text_embeds = model.embed_text(["a dog running on the beach"])
     image_embeds = model.embed_image([Image.open("photo.jpg")])
     similarity = text_embeds @ image_embeds.T
@@ -20,7 +21,7 @@ from .projectors import AudioProjector, VisionProjector
 from .video import embed_video, sample_frames
 
 __all__ = [
-    "GelatoLocal",
+    "QuadEmbed",
     "TextEncoder",
     "VisionEncoder",
     "AudioEncoder",
@@ -31,7 +32,7 @@ __all__ = [
 ]
 
 
-class GelatoLocal:
+class QuadEmbed:
     """Loads all three frozen encoders plus the trained projectors, and
     exposes one `embed_*` method per modality, each returning an L2-normalized
     [batch, 768] tensor in the shared space -- so any two outputs from any