Text-to-Image Model Releases Collection One dataset per launch. A new image model ships, we put it in front of thousands of people, and publish how it actually did against other SOTA models. • 15 items • Updated Sep 22 • 2
Video Model Releases Collection One dataset per launch, text-to-video and image-to-video. Veo, Sora, Kling, Pika, Seedance, ranked by real humans. • 16 items • Updated Sep 22 • 1
Benchmark Release Data Collection The vote data behind every live benchmark on benchmark.ai. One dataset per benchmark. • 7 items • Updated about 1 month ago • 8
Rapidata Benchmarks Collection Generative models ranked by real people. • 8 items • Updated Oct 15 • 4
Video generation: human preferences & benchmarks Collection Human judgements of text-to-video and image-to-video models. Start with the consolidated 823k-vote set. • 32 items • Updated 16 days ago • 1
Image generation: human preferences & benchmarks Collection Text-to-image, SVG and face-generation preference data behind benchmark.ai/image and benchmark.ai/svg. • 37 items • Updated 16 days ago • 1
3D generation: human preferences & benchmarks Collection Image-to-3D and LLM-written 3D scenes judged by people. • 3 items • Updated 16 days ago • 1
Audio & speech: human preferences Collection Text-to-audio / TTS preference judgements (naturalness, friendliness) and audio association studies. • 2 items • Updated 16 days ago • 1
Text & LLM: human evaluation Collection Human judgements of LLM output: translation quality across 10+ languages and multilingual humour. • 4 items • Updated 16 days ago • 1