deepseek-v3.1-terminus-thinking
71.4%
no interval published
aiqbench · LLM benchmarks
71.4%
no interval published
66.2%
no interval published
64.7%
no interval published
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 | |||||
|---|---|---|---|---|---|---|---|---|---|
| 45 | deepseek-v3.1-terminus-thinking | DeepSeek | Oct 2026 | — | 71.4% | — | #97–137 | ||
| 72 | deepseek-v3.2-thinking | DeepSeek | Oct 2026 | — | 66.2% | — | #103–344 | ||
| 82 | deepseek-v3.1-terminus | DeepSeek | Sep 2025 | $0.27 / $1 | 64.7% | — | #55–168 | ||
| 87 | deepseek-v3.2-speciale | DeepSeek | — | — | 64.1% | — | — | ||
| 93 | deepseek-v3.2-exp-thinking | DeepSeek | Oct 2026 | — | 63.1% | — | #77–180 | ||
| 95 | deepseek-v3-0324 | DeepSeek | Mar 2025 | — | 62.8% | — | #137–401 | ||
| 107 | deepseek-v3 | DeepSeek | Dec 2024 | — | 60.5% | — | #178–644 | ||
| 133 | deepseek-v3.2-exp | DeepSeek | Sep 2025 | $0.27 / $0.41 | 55.5% | — | #92–334 |
Showing LiveBench exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.