gpt-5.4-high
1,484±3
1,481CI1,487
aiqbench · LLM benchmarks
Benchmarks
- Sources
- 8
- Entries tracked
- 22,310
- Last sync
- 46 min46 minutes ago
All sources syncing normally
- Arena · Text· 1 hour ago
- Arena · WebDev· 1 hour ago
- Arena · Vision· 1 hour ago
- Arena · Agent· 1 hour ago
- GPQA Diamond· 46 minutes ago
- SWE-bench· 46 minutes ago
- Terminal-Bench· 46 minutes ago
- ARC-AGI-2· 46 minutes ago
Filter by org:
Released:
gpt-5.4-mini-high
1,447±3
1,444CI1,450
gpt-5.4-nano-high
1,417±3
1,414CI1,420
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-5.4-high | openai | Sep 2026 | $2.50 / $15 | 1,484 | 1,481–1,487 | #12–29 | ||
| 67 | gpt-5.4-mini-high | openai | Oct 2026 | $0.75 / $4.50 | 1,447 | 1,444–1,450 | #75–180 | ||
| 127 | gpt-5.4-nano-high | openai | Oct 2026 | $0.20 / $1.25 | 1,417 | 1,414–1,420 | #95–222 |
Showing Factuality Arena exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.