GPT-5.6 Luna (High) vs GPT-6.1 Sol (Low)
Which is better for the money: GPT-5.6 Luna (High) by OpenAI or GPT-6.1 Sol (Low) 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-6.1 Sol (Low) scores 10.0 points higher (42.1 vs 32.1) but costs 2.3× as much to run the benchmark ($250 vs $108; both mid-range class). Pick GPT-6.1 Sol (Low) when the extra intelligence is worth the premium, otherwise GPT-5.6 Luna (High).
| Metric | GPT-5.6 Luna (High) | GPT-6.1 Sol (Low) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 32.1 | 42.1 | +31.2% GPT-6.1 Sol (Low) is 31.2% more intelligent |
| Total AA cost (lower is better) | $108.15 | $250.48 | +131.6% GPT-6.1 Sol (Low) is 2.3× more expensive |
| End-to-end time (lower is better) | 16.71s | 12.91s | −22.7% GPT-6.1 Sol (Low) is 22.7% faster |
| Price per 1M tokens (in / out) | $0.20 / $1.20 | $2.00 / $10.0 | |
| Input modalities | Text, Image | Text, Image |
GPT-6.1 Sol (Low) vs GPT-5.6 Luna (High): 31.2% more intelligence, 131.6% more cost to run the AA benchmark, and 22.7% faster end-to-end.
Today's Pareto frontier
Against the frontier
- GPT-5.6 Luna (High) is off the frontier: at the same or lower cost, Claude Haiku 5.5 (High) scores 37.80 for $94.1.
- GPT-6.1 Sol (Low) is off the frontier: at the same or lower cost, MiMo-V2.6-Pro scores 46.30 for $207.
GPT-5.6 Luna (High)GPT-6.1 Sol (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/gpt-5-6-luna-high/gpt-6-1-sol-low · Markdown: /api/v1/compare/gpt-5-6-luna-high/gpt-6-1-sol-low?format=md · llms.txt · agent skill