Mistral Large 3open
Mistral
35.9#297 of 427 overall
- Input / output
- US$ 0,50 / US$ 1,50 US$ per 1M tokens
- Context
- — tokens
- Release
- 12/02/2025 release date
- Weights
- Open
- Speed
- 0 tok/s first token in 0.00 s
Family history
The 5 versions of this family, from first to newest, by release date. Click a point to open that version.
See the data as a table
| Model | Released | Score | Rank | AA Intel. | Elo Arena |
|---|---|---|---|---|---|
| Mistral Large 2 | 11/18/2024 | 17.4 | #416 | — | 1266 |
| Mistral Large 3 2512 | 12/01/2025 | 17.5 | #414 | — | 1265 |
| Mistral Large 3 | 12/02/2025 | 35.9 | #297 | — | 1428 |
| Mistral Large 4 | 10/06/2026 | 43.9 | #209 | — | — |
| Mistral Large 4 | 10/29/2026 | 42.7 | #227 | 26.0 | 1430 |
Scores by source
Scores by sourcepercentile = % of ranked models it beats| Metric | Value | Percentile | Source |
|---|---|---|---|
| Arena text (Elo) weight 2 | 1428 | 61% | LMArena |
| Arena WebDev (Elo) weight 1 | 1233 | 6% | LMArena |
| Arena vision (Elo) weight 0.5 | 1220 | 37% | LMArena |
| Humanity's Last Exam weight 1 | 4.2 | 15% | Artificial Analysis |
| Terminal-Bench weight 1 | 0.0 | 22% | Artificial Analysis |
| GPQA Diamond | 68.0 | 47% | Artificial Analysis |
| Long-context reasoning | 36.0 | 36% | Artificial Analysis |
By category
| Category | Score | Rank | Metrics |
|---|---|---|---|
| Agentic | 36.0 | #120 | Terminal-Bench |
| Coding | 28.0 | #112 | Arena WebDev (Elo) |
| Reasoning | 37.4 | #275 | Humanity's Last Exam, GPQA Diamond |
| Research | 43.0 | #226 | Long-context reasoning |
| Design | 28.0 | #111 | Arena WebDev (Elo) |
| Chat | 55.5 | #78 | Arena text (Elo) |
| Vision | 43.5 | #67 | Arena vision (Elo) |
History
RankingIA score on each collection day (hover to see the rank).
Daily collection started on 10/30/2026. The chart appears from the second day the model is ranked.