GPT-6 Astra scores 70.5 on the Respan Index (#2), ahead of Muse Spark 1.3 at 54.1 (#13). GPT-6 Astra leads on 14 of the 19 benchmarks both report. Muse Spark 1.3 is 10x 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
| LiveBench Agentic Coding (LiveBench) |
| 57.3% |
| 64.1% |
| LiveBench Coding (LiveBench) | 80.4% | 81.1% |
| LiveBench Data Analysis (LiveBench) | 83% | 79.6% |
| LiveBench Instruction Following (LiveBench) | 75.6% | 78% |
| LiveBench Language (LiveBench) | 89.4% | 82.8% |
| LiveBench Mathematics (LiveBench) | 96.8% | 95.9% |
| LiveBench Reasoning (LiveBench) | 92.7% | 89.7% |
| Mystery Game Puzzles (Epoch AI) | 84% | 25% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 99.2% |
| Terminal-Bench 4.0 (Vals AI) | 59.6% | 24.7% |
| LMArena Agent (LMArena) | 0.1227 | 0.04 |
| LMArena Elo | 1477 | 1494 |
| LMArena Vision (LMArena) | 1284 | 1290 |
| LMArena WebDev (LMArena) | 1788 | 1657 |
| SimpleBench (SimpleBench) | 86.5% | 81.8% |
Higher is better. Each score links to where it was published.