Benchmark
RISEBench - Logical
Scores are percentages.
- Models measured
- 12
- Scores recorded
- 14
- Sources
- 1
- IQ weight
- 0.84
Score distribution
- 11.2%
- Mean
- 7.1%
- Median
- 11.8%
- Std dev
- 1.2%
- 10th percentile
- 34.1%
- 90th percentile
- 0%
- Min
This benchmark has a weight of 0.84 for image-generation models, measured over 10 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro Image· RISEBench data | 37.6% |
| 2 | GPT Image 2· RISEBench data | 34.1% |
| 3 | GPT Image 1.5· RISEBench data | 21.2% |
| 4 | Gemini 2.5 Flash Image· RISEBench data | 18.8% |
| 5 | GPT Image 1· RISEBench data | 10.6% |
| 6 | GPT-Image-1 mini· RISEBench data | 9.4% |
| 7 | Doubao Seedream 4-0· RISEBench data | 7.1% |
| 8 | Gemini 2.0 Flash Experimental· RISEBench data | 4.7% |
| 9 | Qwen Image Edit2511· RISEBench data | 4.7% |
| 10 | Step1X-Edit· RISEBench data | 3.5% |
| 11 | Qwen Image Edit2509· RISEBench data | 2.4% |
| 12 | BAGELw/ CoT· RISEBench data | 1.2% |
| 13 | Flux Kontext Dev· RISEBench data | 1.2% |
| 14 | BAGEL· RISEBench data | 0% |