GPT-5.6 Luna (Max) vs MiMo-V2.5
Which is better for the money: GPT-5.6 Luna (Max) by OpenAI or MiMo-V2.5 by Xiaomi? 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-5.6 Luna (Max) scores 15.5 points higher (37.3 vs 21.8) but costs 5.2× as much to run the benchmark ($320 vs $61.9; both mid-range class). They are not alternatives at one budget: pick GPT-5.6 Luna (Max) when you need intelligence above 21.8, and MiMo-V2.5 when 21.8 is enough.
| Metric | GPT-5.6 Luna (Max) | MiMo-V2.5 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 37.3 | 21.8 | −41.6% MiMo-V2.5 is 41.6% less intelligent |
| Total AA cost (lower is better) | $319.93 | $61.93 | −80.6% MiMo-V2.5 is 80.6% cheaper |
| End-to-end time (lower is better) | 114.07s | 66.80s | −41.4% MiMo-V2.5 is 41.4% faster |
| Price per 1M tokens (in / out) | $0.20 / $1.20 | $0.13 / $0.27 | |
| Input modalities | Text, Image | Text, Image |
MiMo-V2.5 vs GPT-5.6 Luna (Max): 41.6% less intelligence, 80.6% less cost to run the AA benchmark, and 41.4% faster end-to-end.
Today's Pareto frontier
Against the frontier
- GPT-5.6 Luna (Max) is off the frontier: at the same or lower cost, MiMo-V2.6-Pro scores 46.30 for $207.
- MiMo-V2.5 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (Medium) scores 34.50 for $56.4.
GPT-5.6 Luna (Max)MiMo-V2.5Today's Pareto frontierOpen-weights 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/mimo-v2-5-0424 · Markdown: /api/v1/compare/gpt-5-6-luna/mimo-v2-5-0424?format=md · llms.txt · agent skill