Mistral AI
Mistral Large 3
A 675 billion parameter open-source Mixture-of-Experts model ranked highly on community leaderboards.
Our editorial assessment as of Apr 24, 2026, based on the public sources listed below.
Overall rank
#30
Tier
A
Score
78.0
Main category
Overall
Strengths
- A top-performing open-source model.
- Built on a Mixture-of-Experts (MoE) architecture for efficiency.
- Strong ranking on the LMArena leaderboard.
Considerations
- Specific benchmark scores and pricing are not detailed in the provided sources.
- May lag behind the latest proprietary models in raw performance.
Price
Pricing not specified
Score breakdown
Performance
—
Reputation
—
Price
60.0
Recency
—
Verification status
Each field carries its own re-verification date. Amber means older than 90 days — treat those figures as stale until re-checked.
Sources
- TeamDay.ai Frontier AI Models 2026Accessed: 2026-04-20
- AI Index Report 2026Accessed: 2026-04-20