Qwen3.8-Max scores 50.3 on the Respan Index (#22), ahead of DeepSeek V4 Pro at 46.9 (#29). Qwen3.8-Max leads on 12 of the 21 benchmarks both report. DeepSeek V4 Pro is 1.5x cheaper per token at list price.
* Estimated: no published scores in that category. How the Respan Index works
| 45.3% |
| 74.7% |
| GPQA Diamond | 90.9% | 92.6% |
| LiveBench Agentic Coding (LiveBench) | 54.9% | 64.6% |
| LiveBench Coding (LiveBench) | 77.2% | 72.9% |
| LiveBench Data Analysis (LiveBench) | 79.2% | 78.4% |
| LiveBench Instruction Following (LiveBench) | 67.7% | 74.1% |
| LiveBench Language (LiveBench) | 82.1% | 79.7% |
| LiveBench Mathematics (LiveBench) | 95.1% | 91.3% |
| LiveBench Reasoning (LiveBench) | 85.8% | 88.2% |
| Mystery Game Puzzles (Epoch AI) | 43% | 38% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 96.7% | 99.4% |
| SimpleQA Verified | 47% | 45.8% |
| Terminal-Bench 4.0 (Vals AI) | 14.1% | 34.3% |
| LMArena Agent (LMArena) | 0.0119 | 0.0251 |
| LMArena Elo | 1451 | 1481 |
| LMArena WebDev (LMArena) | 1583 | 1671 |
| Terminal-Bench 2.1 | 87.9% | 86.6% |
Higher is better. Each score links to where it was published.