GPT-5.6 Luna (High) vs Mistral Large 3
Which is better for the money: GPT-5.6 Luna (High) by OpenAI or Mistral Large 3 by Mistral? 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 22.8 points higher (32.1 vs 9.3) but costs 2.3× as much to run the benchmark ($108 vs $46.4; mid-range vs budget cost classes). Pick GPT-5.6 Luna (High) when the extra intelligence is worth the premium, otherwise Mistral Large 3.
| Metric | GPT-5.6 Luna (High) | Mistral Large 3 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 32.1 | 9.3 | −71.0% Mistral Large 3 is 71.0% less intelligent |
| Total AA cost (lower is better) | $108.15 | $46.41 | −57.1% Mistral Large 3 is 57.1% cheaper |
| End-to-end time (lower is better) | 12.33s | 7.23s | −41.4% Mistral Large 3 is 41.4% faster |
| Price per 1M tokens (in / out) | $0.20 / $1.20 | $0.50 / $1.50 | |
| Input modalities | Text, Image | Text, Image |
Mistral Large 3 vs GPT-5.6 Luna (High): 71.0% less intelligence, 57.1% 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 (High) is off the frontier: at the same or lower cost, Claude Haiku 5.5 (High) scores 37.80 for $94.1.
- Mistral Large 3 is off the frontier: at the same or lower cost, GPT-6 Luna (Medium) scores 29.90 for $31.2.
GPT-5.6 Luna (High)Mistral Large 3Today'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/mistral-large-3 · Markdown: /api/v1/compare/gpt-5-6-luna-high/mistral-large-3?format=md · llms.txt · agent skill