Claude Opus 5 (Medium) vs GPT-6 Astra (High)
Which is better for the money: Claude Opus 5 (Medium) by Anthropic or GPT-6 Astra (High) by OpenAI? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-25.
Trade-off
Verdict
Depends on your budget
GPT-6 Astra (High) scores 6.1 points higher (50.9 vs 44.8) but costs 1.1× as much to run the benchmark ($2925 vs $2732; both flagship class). Pick GPT-6 Astra (High) when the extra intelligence is worth the premium, otherwise Claude Opus 5 (Medium).
| Metric | Claude Opus 5 (Medium) | GPT-6 Astra (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 44.8 | 50.9 | +13.6% GPT-6 Astra (High) is 13.6% more intelligent |
| Total AA cost (lower is better) | $2,731.91 | $2,925.01 | +7.1% GPT-6 Astra (High) is 7.1% more expensive |
| End-to-end time (lower is better) | 12.85s | 49.71s | +286.8% GPT-6 Astra (High) is 286.8% slower |
| Price per 1M tokens (in / out) | $5.00 / $25.0 | $10.0 / $50.0 | |
| Input modalities | Text, Image | Text, Image |
GPT-6 Astra (High) vs Claude Opus 5 (Medium): 13.6% more intelligence, 7.1% more cost to run the AA benchmark, and 286.8% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Claude Opus 5 (Medium) is off the frontier: at the same or lower cost, Claude Opus 5.5 (High, Default Fallback) scores 53.60 for $2172.
- GPT-6 Astra (High) is off the frontier: at the same or lower cost, Claude Opus 5.5 (High, Default Fallback) scores 53.60 for $2172.
Claude Opus 5 (Medium)GPT-6 Astra (High)Today's Pareto frontierEvery other model
Matchup
Compare something else
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/claude-opus-5-medium/gpt-6-astra-high · Markdown: /api/v1/compare/claude-opus-5-medium/gpt-6-astra-high?format=md · llms.txt · agent skill