Decision support
Kling 2.1 Master vs Ray2
Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.
Kling
Luma AI
Benchmark results
Head to head on 2 benchmarks both models report
- Kling 2.1 Master2leads
- Ray20leads
Arena Text to Video
Arena rating · Higher is better
- Kling 2.1 MasterBest1163+97 ahead · Arena #42CI 1153 to 1173
- Ray21066Arena #48CI 1048 to 1083
Runs and sources
Kling 2.1 Master- default116395% CI 1153 to 1173 · 14,104 votes
Ray2- default106695% CI 1048 to 1083 · 5,250 votes
Arena Image to Video
Arena rating · Higher is better
- Kling 2.1 MasterBest1236+126 ahead · Arena #36CI 1228 to 1244
- Ray21110Arena #48CI 1093 to 1126
Runs and sources
Kling 2.1 Master- default123695% CI 1228 to 1244 · 29,824 votes
Ray2- default111095% CI 1093 to 1126 · 9,521 votes
Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.
Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.
Model specifications
API pricing · USD per 1M tokens
Capabilities and modalities
Architecture and availability
A dash means the field is not reported. Provider-specific performance and pricing may differ.