GPT-6 Astra scores 70.5 on the Respan Index (#2), ahead of Kimi K2.7 Code at 34.1 (#58). GPT-6 Astra leads on 19 of the 19 benchmarks both report. Kimi K2.7 Code is 12x cheaper per token at list price. GPT-6 Astra has the larger context window (1.1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 93.7% |
| 54% |
| GPQA Diamond | 96% | 87.9% |
| LiveBench Agentic Coding (LiveBench) | 57.3% | 45.7% |
| LiveBench Coding (LiveBench) | 80.4% | 74% |
| LiveBench Data Analysis (LiveBench) | 83% | 62.7% |
| LiveBench Instruction Following (LiveBench) | 75.6% | 56.3% |
| LiveBench Language (LiveBench) | 89.4% | 77.9% |
| LiveBench Mathematics (LiveBench) | 96.8% | 79.6% |
| LiveBench Reasoning (LiveBench) | 92.7% | 82.8% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 95.6% |
| SimpleQA Verified | 75.6% | 36.5% |
| Vending-Bench 2 (Andon Labs) | 15514.7 | 5082.94 |
| WeirdML (Håvard Tveit Ihle) | 93.6% | 54.1% |
| LMArena WebDev (LMArena) | 1788 | 1473 |
| SimpleBench (SimpleBench) | 86.5% | 57.9% |
Higher is better. Each score links to where it was published.