Claude Haiku 5.5 (Xhigh) vs GPT-6 Luna (Low)
Which is better for the money: Claude Haiku 5.5 (Xhigh) by Anthropic or GPT-6 Luna (Low) by OpenAI? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-27.
Trade-off
Verdict
Depends on your budget
Claude Haiku 5.5 (Xhigh) scores 19.7 points higher (41.2 vs 21.5) but costs 15× as much to run the benchmark ($158 vs $10.6; mid-range vs budget cost classes). They are not alternatives at one budget: pick Claude Haiku 5.5 (Xhigh) when you need intelligence above 21.5, and GPT-6 Luna (Low) when 21.5 is enough.
| Metric | Claude Haiku 5.5 (Xhigh) | GPT-6 Luna (Low) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 41.2 | 21.5 | −47.8% GPT-6 Luna (Low) is 47.8% less intelligent |
| Total AA cost (lower is better) | $157.58 | $10.63 | −93.3% GPT-6 Luna (Low) is 93.3% cheaper |
| End-to-end time (lower is better) | 69.98s | 6.99s | −90.0% GPT-6 Luna (Low) is 90.0% faster |
| Price per 1M tokens (in / out) | $0.10 / $0.50 | $0.10 / $0.50 | |
| Input modalities | Text, Image | Text, Image |
GPT-6 Luna (Low) vs Claude Haiku 5.5 (Xhigh): 47.8% less intelligence, 93.3% less cost to run the AA benchmark, and 90.0% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Claude Haiku 5.5 (Xhigh) is on today's frontier: nothing is both smarter and cheaper.
- GPT-6 Luna (Low) is on today's frontier: nothing is both smarter and cheaper.
Claude Haiku 5.5 (Xhigh)GPT-6 Luna (Low)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-xhigh/gpt-6-luna-low · Markdown: /api/v1/compare/claude-haiku-5-5-xhigh/gpt-6-luna-low?format=md · llms.txt · agent skill