Benchmark
Sayf AthenaBench RMS F1
Vision benchmarks measure how well models understand images. Higher scores are better.
- Models measured
- 6
- Scores recorded
- 6
- Sources
- 1
- IQ weight
- 1.06
Score distribution
- 0.232
- Mean
- 0.215
- Median
- 0.21
- Std dev
- 0.034
- 10th percentile
- 0.596
- 90th percentile
- 0.0228
- Min
This benchmark has a weight of 1.06 for chat models, measured over 6 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 | Claude Sonnet 4.6rms· Sayf cybersecurity evaluations data docs | 0.596 |
| 2 | GPT 5.4rms· Sayf cybersecurity evaluations data docs | 0.414 |
| 3 | Gemma 4 31B ITrms· Sayf cybersecurity evaluations data docs | 0.215 |
| 4 | Llama 3.3 70B Instructrms· Sayf cybersecurity evaluations data docs | 0.109 |
| 5 | Qwen3.6 35B A3Brms· Sayf cybersecurity evaluations data docs | 0.0341 |
| 6 | GPT OSS 20brms· Sayf cybersecurity evaluations data docs | 0.0228 |
Data credits: Sayf-eval. Copyright (c) 2026 Qatar Computing Research Institute (QCRI). MIT License. (licence)