| Claude Opus 4.8 | GPT-6 Astra | |
|---|---|---|
| CTO Index | 52.8 (#29) | 87.2 (#6) |
| Provider | Anthropic | OpenAI |
| Weights | Closed | Closed |
| Price $/M in / out | $5 / $25 | $10 / $50 |
| Context window | 1M | 1.1M |
| License | Proprietary | Proprietary |
| Terminal-Bench 4.1 | 23.6% | 60.3% |
| WebDev Arena | 1555 | 1764 |
| ARC-AGI-2 | 72.1% | 95.0% |
| Text Arena | 1482 | 1475 |
| ARC-AGI-3 | 1.5% | 62.7% |
Snapshot 22 November 2026 · Scores copied from each benchmark's own leaderboard · How the CTO Index works
Frequently asked questions
Is Claude Opus 4.8 better than GPT-6 Astra?
GPT-6 Astra scores higher on the CTO Index (87.2 vs 52.8). Across the 5 benchmarks both have results, Claude Opus 4.8 wins 1 (Text Arena) and GPT-6 Astra wins 4 (Terminal-Bench, WebDev, ARC-AGI-2, ARC-AGI-3). Claude Opus 4.8 costs 50% less per blended token.
Which is better for coding agents, Claude Opus 4.8 or GPT-6 Astra?
On Terminal-Bench 4.1, GPT-6 Astra scores 60.3% against Claude Opus 4.8's 23.6%. Differences under about three points are within the error bars.
How do Claude Opus 4.8 and GPT-6 Astra compare on price?
Claude Opus 4.8 lists at $5 / $25 and GPT-6 Astra at $10 / $50 per million input / output tokens.