DeepSeek R1 0528 (May '25)
84.9%
no interval published
aiqbench · LLM benchmarks
84.9%
no interval published
84.4%
no interval published
79.5%
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 | |||||
|---|---|---|---|---|---|---|---|---|---|
| 27 | DeepSeek R1 0528 (May '25) | deepseek | May 2025 | $0.50 / $2.15 | 84.9% | — | #89–250 | ||
| 30 | DeepSeek R1 (Jan '25) | deepseek | Jan 2025 | $0.70 / $2.50 | 84.4% | — | #117–389 | ||
| 95 | DeepSeek R1 Distill Llama 70B | deepseek | Jan 2025 | $0.80 / $0.80 | 79.5% | — | #286–643 | ||
| 140 | DeepSeek R1 Distill Qwen 14B | deepseek | — | — | 74.0% | — | #347–812 | ||
| 141 | DeepSeek R1 0528 Qwen3 8B | deepseek | — | — | 73.9% | — | #465–920 | ||
| 142 | DeepSeek R1 Distill Qwen 32B | deepseek | — | — | 73.9% | — | #255–778 | ||
| 212 | DeepSeek R1 Distill Llama 8B | deepseek | — | — | 54.3% | — | — | ||
| 259 | DeepSeek R1 Distill Qwen 1.5B | deepseek | — | — | 26.9% | — | #407–1029 |
Showing MMLU-Pro (Artificial Analysis) exactly as the source published it — its own score and confidence interval, no blending. Select a second axis to build a weighted composite.