Add remaining release files

CHANGELOG.md

@@ -0,0 +1,9 @@
+# Changelog
+
+## 1.0 (2026-10-20)
+
+Initial release: 145 fonts (one pinned word each), 100 colors, 200
+single-instruction edits on shared starting designs. Scoring: font IoU against
+programmatic references with per-case noise floors; max per-channel color
+error; zero-tolerance whole-image change for edits, with an independent
+reviewer's acceptance rate reported alongside.

LICENSE

@@ -0,0 +1,31 @@
+Creative Commons Attribution 4.0 International (CC BY 4.0)
+
+This dataset (cases, prompts, starting designs, assets, model outputs, judge
+verdicts) is licensed under CC BY 4.0:
+https://creativecommons.org/licenses/by/4.0/
+
+The scoring code under scoring/ is licensed under the MIT License:
+
+MIT License
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+Font files under data/font_cache/ are Google Fonts, each under its own license
+(OFL/Apache/UFL) as published in the google/fonts repository at the pinned
+commit recorded in data/fonts.jsonl.

README.md

@@ -1,3 +1,137 @@
 ---
-license: mit
+license: cc-by-4.0
+task_categories:
+  - text-to-image
+  - image-to-image
+language:
+  - en
+tags:
+  - benchmark
+  - design
+  - typography
+  - color
+  - image-editing
+pretty_name: "VybBench 1.0: Design Fidelity Benchmark"
 ---
+
+# VybBench 1.0
+**Almost isn't good enough.** VybBench measures how well models maintain
+design fidelity by comparing how effectively they generate specific
+fonts and colors, and perform non-destructive edits.
+
+Our dataset consists of 145 fonts, 100 colors, and 200 edits. We have each model
+perform each task three times and score each generation individually.
+
+We compare top image generation models: 
+ChatGPT Image (gpt-image-2.5-sunburst), Meta Muse, and Recraft v4.1. We also report
+the results for [Vyb 1.0](https://cocolabs.ai/vyb).
+
+Full Disclosure: VybBench was built by Coco Labs, the makers of Vyb. The full
+methodology and rationale are in
+[the launch post](https://cocolabs.ai/blog/introducing-vybbench).
+
+## Results
+
+### Font fidelity
+We sample 145 [Google Fonts](https://fonts.google.com). For every font
+we choose a word to generate. We generate that word programmatically
+as well as through model calls. We compute the IoU of the ink between
+the ground truth generation and each model generation. The higher the 
+score the better. As each model is allowed
+to scale and place the text individually, we compute the noise floor by
+measuring the IoU of multiple ground truth generations with the same font.
+
+
+| Model | Mean | Worst 5% of cases |
+|---|---|---|
+| **Vyb 1.0** | **0.90** | **0.78** |
+| ChatGPT Image (`gpt-image-2.5-sunburst`) | 0.76 | 0.28 |
+| Muse (`meta/muse-image`) | 0.51 | 0.18 |
+| Recraft (`v4.1`) | 0.46 | 0.14 |
+| *Measurement noise floor* | *0.94* | *0.88* |
+
+![IoU scores for generating Hazel in Playwrite BE WAL](figures/bench-text.png)
+
+*An example from the benchmark. The ground truth is shown as is each model's
+output. Only Vyb generates an output that is close to the ground truth.*
+
+### Color fidelity
+To measure color fidelity, we pick 100 different colors. We ask each model to generate
+a rectangle in the specified color, given as a hex code. We then calculate the
+per-channel error for each of R, G, and B (out of 255) and report the maximum across
+channels.
+
+
+| Model | Mean | Worst 5% of cases |
+|---|---|---|
+| **Vyb 1.0** | **0.0** | **0.0** |
+| ChatGPT Image (`gpt-image-2.5-sunburst`) | 3.9 | 8 |
+| Muse (`meta/muse-image`) | 5.7 | 10 |
+| Recraft (`v4.1`) | 27.2 | 56 |
+
+Here we see that only Vyb reproduces every requested color exactly.
+
+### Edit fidelity
+
+We provide every model the same starting design and the same edit prompt.
+For Vyb, the design is provided in the layered SVG format it accepts. For
+the leading image generation models, the SVG is rendered and provided in
+the PNG format they accept. 
+
+For each model generation, we measure the fraction of pixels in the image
+that were changed and have an LLM judge score whether the edit was performed
+(acceptance rate).
+
+We report the fraction of pixels changed (lower is better) and acceptance
+rate (higher is better).
+
+| Model | Pixels changed, mean | Worst 5% of cases | Acceptance rate |
+|---|---|---|---|
+| **Vyb 1.0** | **7.3%** | **31.4%** | 99.2% |
+| ChatGPT Image (`gpt-image-2.5-sunburst`) | 95.5% | 99.7% | **100%** |
+| Muse (`meta/muse-image`) | 91.9% | 99.6% | 98.0% |
+| Recraft (`v4.1`) | 98.9% | 100% | 51.2% |
+
+
+![Edit example of changing text color](figures/bench-edit.png)
+
+*An example of an edit where the request is to change the color
+of the "30% OFF" text to coral pink. Models other than Vyb change unrelated parts of the image.*
+
+## Reproduce
+
+To judge new outputs, you need a Gemini API key.
+
+```
+pip install google-genai
+GEMINI_API_KEY=... python -m scoring.judge
+```
+
+You can rescore by running:
+
+```
+pip install numpy pillow scikit-image
+python -m scoring
+```
+
+All designs are synthetic. They contain fictional brands with AI-generated
+imagery.
+
+Run: October 2026.
+
+## Add your model
+
+We are happy to include your model. Create a discussion on this dataset with
+your model's outputs (follow the `outputs/` layout, provide three attempts per
+case).
+
+## Citation
+
+```bibtex
+@misc{vybbench2026,
+  title = {VybBench 1.0: A Design Fidelity Benchmark},
+  author = {{Coco Labs}},
+  year = {2026},
+  url = {https://huggingface.co/datasets/cocolabs/vybbench}
+}
+```

prompts/color.txt

@@ -0,0 +1 @@
+A plain white (#FFFFFF) image with a single solid rectangle of {hex} in the middle. No border, no shadow, no texture, no gradient.

prompts/font.txt

@@ -0,0 +1 @@
+A plain white image. The word "{word}" in solid black, centered, in the font {family} {variant}. Nothing else in the image.

prompts/ours_aspect_hint.txt

@@ -0,0 +1,2 @@
+Appended to the prompt for the Vyb arm only (its API has no size parameter; a stated ratio is its equivalent):
+ (1:1 aspect ratio)

scores.json

@@ -0,0 +1,103 @@
+{
+  "fonts": {
+    "ours": {
+      "mean": 0.9030490100136523,
+      "median": 0.9258799043210836,
+      "p20": 0.8745755467432859,
+      "p10": 0.832307584711683,
+      "p5": 0.7770452071394829
+    },
+    "gpt": {
+      "mean": 0.760261184258689,
+      "median": 0.8357328688053662,
+      "p20": 0.7162267343485618,
+      "p10": 0.4515604183123666,
+      "p5": 0.2788465144185698
+    },
+    "muse": {
+      "mean": 0.5088765844582986,
+      "median": 0.5270965880620131,
+      "p20": 0.30399521489013603,
+      "p10": 0.22172956995061466,
+      "p5": 0.18329736902065374
+    },
+    "recraft": {
+      "mean": 0.4621657925254427,
+      "median": 0.47246518672074095,
+      "p20": 0.27166478160176655,
+      "p10": 0.17692216923835102,
+      "p5": 0.14091297333206798
+    },
+    "noise_floor": {
+      "mean": 0.9441404482758621,
+      "median": 0.951594,
+      "p20": 0.926101,
+      "p10": 0.907312,
+      "p5": 0.879898
+    }
+  },
+  "colors": {
+    "ours": {
+      "mean": 0.0,
+      "median": 0.0,
+      "p80": 0.0,
+      "p90": 0.0,
+      "p95": 0.0
+    },
+    "gpt": {
+      "mean": 3.9366666666666665,
+      "median": 4.0,
+      "p80": 5.0,
+      "p90": 7.0,
+      "p95": 8.0
+    },
+    "muse": {
+      "mean": 5.666666666666667,
+      "median": 5.0,
+      "p80": 7.0,
+      "p90": 9.0,
+      "p95": 10.0
+    },
+    "recraft": {
+      "mean": 27.19047619047619,
+      "median": 23.0,
+      "p80": 40.0,
+      "p90": 48.0,
+      "p95": 56.0
+    }
+  },
+  "edits": {
+    "ours": {
+      "mean": 0.0732679820747017,
+      "median": 0.026280864197530864,
+      "p80": 0.10023619957537155,
+      "p90": 0.2635643325617284,
+      "p95": 0.3138264746227709,
+      "satisfied_rate": 0.9916666666666667
+    },
+    "gpt": {
+      "mean": 0.955290820743786,
+      "median": 0.9798226318881986,
+      "p80": 0.9928648834019205,
+      "p90": 0.9954869684499315,
+      "p95": 0.9973703703703704,
+      "satisfied_rate": 1.0
+    },
+    "muse": {
+      "mean": 0.9185079350255037,
+      "median": 0.9642622599451303,
+      "p80": 0.9883641975308642,
+      "p90": 0.9940493827160494,
+      "p95": 0.9963680997876858,
+      "satisfied_rate": 0.98
+    },
+    "recraft": {
+      "mean": 0.9891732617765999,
+      "median": 0.9999348690414952,
+      "p80": 0.9999986282578875,
+      "p90": 1.0,
+      "p95": 1.0,
+      "satisfied_rate": 0.5116666666666667
+    }
+  }
+}

scoring/__init__.py

@@ -0,0 +1,14 @@
+"""VybBench scoring: standalone, offline, no model APIs.
+
+Every number in the benchmark recomputes from the released artifacts alone:
+
+* fonts  -- IoU against a programmatic (PIL) render of the pinned font file,
+            after crop/scale alignment (``fonts.font_mismatch``).
+* colors -- max per-channel delta of the produced rectangle against the
+            requested hex (``colors.color_error``).
+* edits  -- fraction of pixels changed at all between the shared starting
+            design and the result (``edits.whole_change``), with the recorded
+            LLM reviewer's acceptance rate reported alongside.
+
+``python -m scoring`` recomputes everything from ``outputs/``.
+"""