gpt-5.2-2025-12-11
40.7%±2.9%
37.8%CI43.6%
aiqbench · LLM benchmarks
Benchmarks
- Sources
- 8
- Entries tracked
- 22,318
- Last sync
- 7 min7 minutes ago
All sources syncing normally
- Arena · Text· 8 hours ago
- Arena · WebDev· 8 hours ago
- Arena · Vision· 8 hours ago
- Arena · Agent· 8 hours ago
- GPQA Diamond· 7 minutes ago
- SWE-bench· 7 minutes ago
- Terminal-Bench· 7 minutes ago
- ARC-AGI-2· 7 minutes ago
Filter by org:
Released:
gpt-5-2025-08-07
32.4%±2.8%
29.7%CI35.2%
One grid, every column. Pick axes in the Axes panel — one axis ranks like a single board, several blend into a weighted composite, all of them read as the index. Rank is assigned before any filter and never renumbers.
| Model | Org | ± CI | Swing | Add to compare | |||||
|---|---|---|---|---|---|---|---|---|---|
| 8 | gpt-5.2-2025-12-11 | OpenAI | Dec 2025 | $1.75 / $14 | 40.7% | 37.8%–43.6% | #34–93 | ||
| 16 | gpt-5-2025-08-07 | OpenAI | Aug 2025 | $1.25 / $10 | 32.4% | 29.7%–35.2% | #64–149 |
Showing FrontierMath exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.