GPT-6 Luna (High) vs Mistral Large 4 (Medium)
Which is better for the money: GPT-6 Luna (High) by OpenAI or Mistral Large 4 (Medium) by Mistral? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-30.
Trade-off
Verdict
Depends on your budget
Mistral Large 4 (Medium) scores 1.7 points higher (23.1 vs 21.4) but costs 15× as much to run the benchmark ($936 vs $62.6; premium vs mid-range cost classes). They are not alternatives at one budget: pick Mistral Large 4 (Medium) when you need intelligence above 21.4, and GPT-6 Luna (High) when 21.4 is enough.
| Metric | GPT-6 Luna (High) | Mistral Large 4 (Medium) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 21.4 | 23.1 | +7.9% Mistral Large 4 (Medi… is 7.9% more intelligent |
| Total AA cost (lower is better) | $62.58 | $936.36 | +1396.3% Mistral Large 4 (Medi… is 15× more expensive |
| End-to-end time (lower is better) | 15.20s | Not reported | — not comparable |
| Price per 1M tokens (in / out) | $0.10 / $0.50 | $0.70 / $2.10 | |
| Input modalities | Text, Image | Text, Image |
Mistral Large 4 (Medium) vs GPT-6 Luna (High): 7.9% more intelligence, 1396.3% more cost to run the AA benchmark, and end-to-end time not fully reported.
Today's Pareto frontier
Against the frontier
- GPT-6 Luna (High) is on today's frontier: nothing is both smarter and cheaper.
- Mistral Large 4 (Medium) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Xhigh) scores 52.30 for $718.
GPT-6 Luna (High)Mistral Large 4 (Medium)Today'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-6-luna-high/mistral-large-4-medium · Markdown: /api/v1/compare/gpt-6-luna-high/mistral-large-4-medium?format=md · llms.txt · agent skill