InferBench Text-to-Image - GenAI-Bench — leaderboard
InferBench compares text-to-image inference providers and optimizations on quality metrics such as GenAI-Bench VQA, DrawBench, HPS, and PartiPrompts, alongside speed and price.
Metric: GenAI-Bench VQA. Source: huggingface.co. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | FLUX 2 beta (Black Forest Labs) | 0.97 |
| 2 | FLUX 2 pro (Black Forest Labs) | 0.96 |
| 3 | FLUX Krea (Replicate, go_fast) | 0.82 |
| 4 | FLUX Krea (Together AI, dev) | 0.8 |
| 5 | FLUX Krea (fal.ai) | 0.79 |
| 6 | FLUX Schnell (fal.ai, schnell) | 0.79 |
| 7 | FLUX Schnell (Replicate, speed_mode_juiced) | 0.79 |
| 8 | FLUX Schnell (Prodia, fast) | 0.79 |
| 9 | FLUX 1.1 Pro (Replicate) | 0.78 |
| 10 | FLUX Schnell (Replicate, go_fast) | 0.78 |
Interactive version: theaggregate.ai/benchmark?slug=inferbench-text-to-image-genai-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.