Claude Haiku 5.5 (High) vs GPT-6 Sol (High)
Which is better for the money: Claude Haiku 5.5 (High) by Anthropic or GPT-6 Sol (High) by OpenAI? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-26.
Trade-off
Verdict
Depends on your budget
GPT-6 Sol (High) scores 4.6 points higher (42.4 vs 37.8) but costs 6.4× as much to run the benchmark ($605 vs $94.1; premium vs mid-range cost classes). They are not alternatives at one budget: pick GPT-6 Sol (High) when you need intelligence above 37.8, and Claude Haiku 5.5 (High) when 37.8 is enough.
| Metric | Claude Haiku 5.5 (High) | GPT-6 Sol (High) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 37.8 | 42.4 | +12.2% GPT-6 Sol (High) is 12.2% more intelligent |
| Total AA cost (lower is better) | $94.07 | $604.52 | +542.6% GPT-6 Sol (High) is 6.4× more expensive |
| End-to-end time (lower is better) | 28.27s | 21.24s | −24.9% GPT-6 Sol (High) is 24.9% faster |
| Price per 1M tokens (in / out) | $0.10 / $0.50 | $2.00 / $10.0 | |
| Input modalities | Text, Image | Text, Image |
GPT-6 Sol (High) vs Claude Haiku 5.5 (High): 12.2% more intelligence, 542.6% more cost to run the AA benchmark, and 24.9% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Claude Haiku 5.5 (High) is on today's frontier: nothing is both smarter and cheaper.
- GPT-6 Sol (High) is off the frontier: at the same or lower cost, GPT-6.1 Sol (High) scores 50.20 for $521.
Claude Haiku 5.5 (High)GPT-6 Sol (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-haiku-5-5-high/gpt-6-sol-high · Markdown: /api/v1/compare/claude-haiku-5-5-high/gpt-6-sol-high?format=md · llms.txt · agent skill