Historical Performance - Images
90-day score trends for top image generation AI models. Track composite score changes, model entry dates, and key events that impacted rankings.
Score Trends
| Rank | Model | Provider | Score | Trend | 30d | 60d | 90d |
|---|---|---|---|---|---|---|---|
| #1 | GPT-5.4 Image 2 | OpenAI | 91.3 | -0.7 | -0.7 | -0.7 | |
| #2 | GPT-5 Image Mini | OpenAI | 88.5 | +27.5 | +27.5 | +27.5 | |
| #3 | GPT-5 Image | OpenAI | 88.5 | +28.1 | +28.1 | +28.1 | |
| #4 | Nano Banana (Gemini 2.5 Flash Image) | 76.9 | +37.6 | +37.6 | +37.6 | ||
| #5 | Nano Banana 2 (Gemini 3.1 Flash Image) | 40.0 | 0.0 | 0.0 | 0.0 | ||
| #6 | Nano Banana Pro (Gemini 3 Pro Image) | 40.0 | 0.0 | 0.0 | 0.0 | ||
| #7 | Nano Banana 2 (Gemini 3.1 Flash Image Preview) | 40.0 | -11.8 | -11.8 | -11.8 | ||
| #8 | Nano Banana Pro (Gemini 3 Pro Image Preview) | 40.0 | -4.7 | -4.7 | -4.7 | ||
| #9 | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | 22.4 | 0.0 | 0.0 | 0.0 | ||
| #10 | Stable Diffusion 3.5 | Stability AI | 11.4 | +5.4 | +5.4 | +5.4 |
Model Timeline
| Date Entered | Model | Provider | Entry Rank | Current Rank |
|---|---|---|---|---|
| 2025-11-16 | Nano Banana (Gemini 2.5 Flash Image) | #5 | #4 | |
| 2025-11-11 | GPT-5 Image | OpenAI | #3 | #3 |
| 2025-10-21 | Stable Diffusion 3.5 | Stability AI | #16 | #10 |
| 2025-09-29 | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | #13 | #9 | |
| 2025-09-05 | Nano Banana Pro (Gemini 3 Pro Image) | #7 | #6 | |
| 2025-08-25 | Nano Banana 2 (Gemini 3.1 Flash Image Preview) | #10 | #7 | |
| 2025-07-19 | GPT-5 Image Mini | OpenAI | #6 | #2 |
| 2025-07-08 | Nano Banana Pro (Gemini 3 Pro Image Preview) | #17 | #8 | |
| 2025-07-06 | GPT-5.4 Image 2 | OpenAI | #4 | #1 |
| 2025-07-04 | Nano Banana 2 (Gemini 3.1 Flash Image) | #10 | #5 |
Key Events
Related
We track composite scores for the top image generation AI models over a 90-day rolling window. The scores factor in generation quality, pricing, API availability, and feature capabilities with hourly updates.
Check the Score Trends table above to see 30-day, 60-day, and 90-day score changes for each model. Major version updates like new DALL-E or Stable Diffusion releases often cause significant score shifts.
The historical trend lines above show how each model's composite score has evolved. While rankings shift with each major release, the competition between these models drives rapid improvements across the category.