Grok 4.5 scores 43.2 on the Respan Index (#38), ahead of Qwen3.8-Flash-Next at 42.7 (#39). Qwen3.8-Flash-Next leads on 6 of the 10 benchmarks both report. Qwen3.8-Flash-Next has the larger context window (1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 71.5% |
| 77.1% |
| LiveBench Language (LiveBench) | 82.8% | 74.6% |
| LiveBench Mathematics (LiveBench) | 90.8% | 85.8% |
| LiveBench Reasoning (LiveBench) | 87.2% | 87.4% |
| LMArena Agent (LMArena) | 0.0121 | -0.0073 |
| LMArena WebDev (LMArena) | 1552 | 1638 |
Higher is better. Each score links to where it was published.