GPT-5.6 Luna (High) vs MiMo-V2.5
Which is better for the money: GPT-5.6 Luna (High) 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 (High) scores 10.3 points higher (32.1 vs 21.8) but costs 1.7× as much to run the benchmark ($108 vs $61.9; both mid-range class). Pick GPT-5.6 Luna (High) when the extra intelligence is worth the premium, otherwise MiMo-V2.5.
| Metric | GPT-5.6 Luna (High) | MiMo-V2.5 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 32.1 | 21.8 | −32.1% MiMo-V2.5 is 32.1% less intelligent |
| Total AA cost (lower is better) | $108.15 | $61.93 | −42.7% MiMo-V2.5 is 42.7% cheaper |
| End-to-end time (lower is better) | 13.41s | 66.80s | +398.1% MiMo-V2.5 is 398.1% slower |
| 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 (High): 32.1% less intelligence, 42.7% less cost to run the AA benchmark, and 398.1% slower 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.
- 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 (High)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-high/mimo-v2-5-0424 · Markdown: /api/v1/compare/gpt-5-6-luna-high/mimo-v2-5-0424?format=md · llms.txt · agent skill