deepseek-v4-pro
1,455±3
1,452CI1,458
aiqbench · LLM benchmarks
1,455±3
1,452CI1,458
1,447±3
1,444CI1,451
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 | |||||
|---|---|---|---|---|---|---|---|---|---|
| 50 | deepseek-v4-pro | deepseek | Aug 2026 | $0.66 / $1.98 | 1,455 | 1,452–1,458 | #17–53 | ||
| 63 | deepseek-v4-pro-high-preview | deepseek | Oct 2026 | — | 1,447 | 1,444–1,451 | #62–161 |
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.