Everything public, as pulled from the Hugging Face datasets API (expand[]=downloadsAllTime), the dataset-viewer size API, the arXiv API, the Semantic Scholar Graph API and GitHub org listings on 2 September 2026. Download and like figures are snapshots and will drift. mo/all = last-month / all-time downloads.
Read this beside What publishing actually bought them, which is what these numbers mean, and What nobody has built, which is what is missing from them.
Contra Labs
Eight datasets, all CC-BY-4.0, all created between 18 May and 17 July 2026, all last modified within five days of creation and none touched since 22 July 2026. Every card except HCB is labelled a "preview release". Every card carries UTM-tagged links back to contralabs.com and a "Working with Contra Labs" section with a partnerships address — these are lead-generation assets with a dataset attached, and they say so.
| Artefact | Created / modified | Licence | Size | mo/all · likes |
|---|---|---|---|---|
contralabs/HumanCreativityBenchmark | 18 May / 24 Jun 2026 | CC-BY-4.0 | 8,012 rows over 5 CSVs — 95 prompts, 380 outputs, 3,174 pairwise, 2,116 scalar, 2,247 qualitative | 278 / 532 · 2 |
contralabs/photoshop-creative-design-trajectories | 2 Jun / 10 Jun 2026 | CC-BY-4.0 | 294 steps, 163 MB | 232 / 822 · 3 |
contralabs/firefly-creative-campaign-trajectories | 9 Jun / 10 Jun 2026 | CC-BY-4.0 | 137 steps, 63 MB | 101 / 399 · 1 |
contralabs/gemini-creative-campaign-trajectories | 9 Jun / 10 Jun 2026 | CC-BY-4.0 | 266 steps, 116 MB | 150 / 432 · 1 |
contralabs/video-detail-annotation | 1 Jul / 2 Jul 2026 | CC-BY-4.0 | 15 rows — 3 video models, timestamped editor comments | 369 / 1,007 · 3 |
contralabs/premiere-video-editing-trajectories | 14 Jul / 16 Jul 2026 | CC-BY-4.0 | 234 steps over 4 trajectories, 182 MB | 614 / 1,415 · 12 |
contralabs/creative-ad-design-dataset | 17 Jul / 22 Jul 2026 | CC-BY-4.0 | 35 rows — brand brief → approved ad, 126 MB | 101 / 146 · 2 |
contralabs/descript-video-editing-trajectories | 17 Jul / 18 Jul 2026 | CC-BY-4.0 | 803 steps, 389 MB | 283 / 380 · 0 |
| The Human Creativity Benchmark, 2606.30561 | 29 Jun 2026, 30 pp | arXiv default | 28 evaluators, 93 prompts, 80 sessions, ~15.5k judgments claimed | 0 citations |
| TASTE, 2605.20731 (with Lica) | v1 20 May, v2 2 Jun 2026 | CC-BY-4.0 | 10 designers, 9 criteria, 80 prompts each, 21,600 pairwise | 0 citations |
| contralabs.com/research | 8 Apr – 26 Aug 2026 | none stated | 39 dated studies + 5 dataset pages + 1 methodology page | n/a |
HF org contralabs | — | — | 8 datasets, 0 models, 0 Spaces, 0 papers, 5 followers | — |
| GitHub | — | — | No organisation exists — /orgs/contralabs and /orgs/contra-hq both 404 | — |
Org totals: 5,133 all-time downloads, 24 likes. Total trajectory volume across the five trajectory repos: 1,769 steps.
Two details worth carrying. The trajectory schema is the strongest thing in the whole inventory — per-step image, thought transcribed from the editor's spoken narration rather than model-synthesised, action_type, structured tool_call, execution_paths (MCP tool / keyboard shortcut / menu path) and a preferred_execution marker (The trajectory moat). And the HCB paper's data link points at huggingface.co/datasets/contra-labs/HCB, which 307-redirects to the live repo — the link works, but the org contra-labs does not exist, so the paper cites a repo name Contra abandoned.
Lica World, now Gamma
| Artefact | Date | Licence | Size | mo/all · likes |
|---|---|---|---|---|
lica-world/GDB | 13 Apr / 5 May 2026 | Apache-2.0 | 33,886 rows, 17.0 GB, 40 task configs | 439 / 4,615 · 1 |
lica-world/lica-dataset | 14 Apr 2026 | CC-BY-4.0 | 1,148 compositions — a 0.07% sample of the 1,550,244 in the paper | 111 / 2,710 · 4 |
lica-world/GDB-video | 1 Jul 2026 | Apache-2.0 | 654 rows over 136 animated layouts; reference/sora2/veo31 configs | 393 / 1,580 · 1 |
purvanshi/TASTE | 28 / 29 May 2026 | MIT | 36,695 rows — 14,460 rankings, 3,200 hallucination flags, 721 prompts, 644 assets, 10 evaluators | 408 / 1,809 · 9 |
| LICA, 2603.16098 | 17 Mar 2026 | — | 1,550,244 compositions, 971,850 templates, 27,261 animated | 5 citations, 4 self |
| GDB, 2604.04192 | 5 Apr 2026 | — | 50 tasks over 5 axes (site says 49) | 4 citations, 3 self |
| SVG structural metrics, 2604.08809 | 9 Apr 2026 | — | 19,000+ edits, 5 edit types, 5 systems | 1 citation (self) |
| Evaluating Design Video Generation, 2605.16223 | 15 May 2026 · ICML 2026 Workshop on Human-AI Co-Creativity | — | 4 evaluation dimensions | 0 citations |
| A Typography Benchmark for Co-Creative Graphic-Design Agents | forthcoming | — | no arXiv id; visible only as a citing edge in Semantic Scholar | — |
purvanshi/lica-bench | — | LICENSE present, type not surfaced | 45 tasks / 39 benchmarks / 7 domains over 1,148 LICA layouts | 10★, 2 forks |
purvanshi-lica/taste | — | MIT | analysis/ + data/ + taste-scorer (pip-installable, Qwen3-VL-Embedding-2B backbone) | 3★, 1 fork |
Note that TASTE and both repositories live on personal accounts, not the org — which matters now that Lica belongs to Gamma. taste-scorer ships code and data and no trained checkpoint.
AfterQuery
Six datasets, 23 followers, seven org members — the largest Hugging Face footprint of the specialist-data companies, and the only corpus in this document with real citation counts.
| Artefact | Date | Licence | Size | mo/all · likes | Citations |
|---|---|---|---|---|---|
AfterQuery/FinanceQA | 29 Jan / 21 Feb 2025 | Apache-2.0 | 148 rows | 187 / 6,311 · 21 | 17 (1 influential) |
AfterQuery/vader | 16 / 23 May 2025 | CC-BY-4.0 | 2,659 files; 174 real vulnerabilities | 286 / 5,040 · 4 | 8 |
AfterQuery/ui-bench | 25 Aug 2025 | none set | 30 client-style briefs, 5 categories | 44 / 395 · 2 | 2 |
AfterQuery/App-Bench | 10 Dec 2025 | none set | 6 rows | 98 / 627 · 4 | — |
AfterQuery/MCP-Universe | 29 Jan 2026 | none set | 2 rows — spreadsheet tasks from Goldman Sachs / Evercore experts | 54 / 549 · 0 | — |
AfterQuery/OmicsBench-grader-databases | 10 Apr / 3 May 2026 | MIT | synonym databases for deterministic biology grading | 46 / 1,623 · 0 | — |
| Market-Bench, 2512.12264 | 13 Dec 2025 | — | 3 strategies, 13 models, 5 rounds | 1 | |
| IDE-Bench, 2601.20886 | 28 Jan 2026 | — | 80 tasks, 8 never-published repos | 2 | |
| SpreadsheetBench 2, 2606.29955 | 29 Jun 2026 | — | 321 tasks, best model 34.89% | 2 (1 influential) | |
| GitHub org, 11 repos | — | mixed | harbor (2★, Apache-2.0, official harness for Terminal-Bench 2.0), vader (11★), FinanceQA (9★), anvil (8★), IDE-Bench (5★), swift-anvil (2★), appbench.ai-docs (1★) | ~38★ total |
MCP-Universe itself is Salesforce AI Research's benchmark (arXiv 2508.14704), not AfterQuery's — the HF repo is a two-row contribution to it. And three of six datasets carry no licence at all, which for a $3.2B-valuation company is a governance gap and a defect worth naming (How an artefact travels). harbor has 1,286 commits and two stars: heavily built, barely adopted, and it contains no creative or design environments.
Design Arena
| Artefact | Detail |
|---|---|
HF org designarena | exists on a paid "team" plan, 3 members, 7 followers, tagline "RL, fine-turning, model alignment, human preference" — and 0 datasets, 0 models, 0 Spaces, 0 papers |
| GitHub org, 7 repos | agent-runner (105★, MIT), html-to-pptx (21★, MIT), MicroEvals (7★), audio-arena-bench (6★), LeWitt-Bench (4★, a Sol LeWitt instruction-art dataset), audio-agent-bench (3★), docs (0★) — ~146★ total |
| Leaderboards | designarena.ai, ~40+ boards, Elo / Bradley–Terry, continuously updated from public votes. No vote total, no downloadable eval set, no public dataset |
| notes.designarena.ai | 12 posts, 23 Feb – 4 Aug 2026, including Fullstack Arena Methodology and the Audio Realism Benchmark |
| arXiv | none, under any of Design Arena / Intelligence / Arcada Labs |
A paid team org that is completely empty is the single most legible negative in the inventory. It is consistent with two readings — a distribution plan provisioned and not yet executed, or preference data being distributed privately to buyers — and the record does not separate them (What this plan could not establish). Note also that Design Arena has open-sourced tools and never open-sourced preference data: html-to-pptx is a HTML-to-PowerPoint conversion library, not a corpus. ArcadaLabs on Hugging Face is an unrelated 2022 account hosting two MLX conversions and is not Design Arena's parent; earlier notes identifying it as such should be corrected. Full read at Design Arena, in full.
Taste Labs
| Artefact | Detail |
|---|---|
| Hugging Face | No org. tastelabs, taste-labs and TasteLabs all 404 |
| GitHub | No organisation found |
| arXiv | No paper |
| Datasets, benchmarks, leaderboards, code | None |
| Blog | Six posts, 16 June – 16 August 2026 |
| Requests for research | 16 Aug 2026, 14-min read, Hamidah Oderinwale — see The eleven problems a rival published |
| Prototype fellowship | 16 Aug 2026 — grants, compute, custom human data, collaborative publishing |
Taste Labs has published no technical artefact of any kind and raised $18.5M co-led by CRV and Amplify. It is nonetheless the more dangerous competitor for reasons that have nothing to do with this table (Taste Labs, in full).
Gamma
gamma-app is a verified Hugging Face org with 3 members, 6 followers and 0 datasets, 0 models, 0 Spaces, 0 papers. Gamma acquired Lica on 25 August 2026 (TechCrunch); lica.world remains live with five blog posts and the GDB explorer, and all three Lica HF repos plus both personal-account repositories are still public. Nothing has been migrated, withdrawn or relicensed. That is a window, not a policy, and it may close.
LMArena, as a ruler
| Artefact | Date | Size | mo/all · likes |
|---|---|---|---|
lmarena-ai/VisionArena-Chat | 14 Jan / 4 Feb 2025 | 230K VLM conversations | 3,504 / 162,688 · 15 |
lmarena-ai/leaderboard-dataset | 2 Apr 2026, last modified 2 Sep 2026 | 2,277,369 rows, 21 configs incl. text_to_image, text_to_video, image_edit, webdev | 25,144 / 126,168 · 23 |
lmarena-ai/search-arena-24k | 15 May 2025 / 3 Mar 2026 | 24k | 465 / 81,199 · 40 |
lmarena-ai/webdev-arena-preference-10k | 15 Jan 2025 | 10k webdev preferences | 348 / 78,838 · 20 |
lmarena-ai/arena-human-preference-140k | 1 Aug 2025 | 140k | 2,568 / 56,505 · 55 |
lmarena-ai/arena-human-preference-55k | 2 May 2024 | 55k | 1,867 / 40,661 · 159 |
lmarena-ai/arena-human-preference-100k | 4 Feb 2025 | 100k | 811 / 26,713 · 49 |
lmarena-ai/Arena-T2I-Hard | 30 Jun 2026 | 310 prompts, ~30 decomposed yes/no constraints each | 84 / 268 · 1 |
| Org totals | 2023 – present | 27 datasets, 75 models, 10 Spaces across lmarena-ai, lmarena, lmsys | arena-leaderboard Space: 4,980 likes |
| GitHub | — | FastChat (40k★), RouteLLM (5.4k★), arena-hard-auto (1.1k★), copilot-arena (364★), p2l (275★), arena-rank (109★), PPE (65★), search-arena (58★) | ~47,000★ |
Arena-T2I-Hard is the strategically interesting one: its own paper finds that high public-arena rankings fail to predict faithfulness — LMArena publishing the critique of its own headline metric and shipping the replacement. See The arena layer.
The reference point that closed a gap
PatronusAI/figmatrace plus Qwen3.8-27B-Figmatrace-SFT, created 19 August and last modified 25 August 2026, CC-BY-4.0: 3,469 design trajectories, 22.3 GB, 200+ hours of expert video, 126 tasks, at 1,784 downloads and 12 likes in a fortnight; the model at 19 downloads (arXiv 2608.21460). It is the largest open design-trajectory corpus in existence and it did not come from anyone in this inventory.
No pricing. Not one organisation here publishes a rate for a preference dataset, a trajectory hour or an evaluation engagement. No downstream usage. Zero models on Hugging Face declare training on any Contra dataset, any Lica dataset, purvanshi/TASTE or AfterQuery/ui-bench; FinanceQA has exactly one. And no aesthetic reward-model weights from anybody — the most conspicuous absence in the whole record, treated at What nobody has built.