Mistral Small 4
Mistral
21.6#396 of 427 overall
- Input / output
- US$ 0,15 / US$ 0,60 US$ per 1M tokens
- Context
- — tokens
- Release
- 03/16/2026 release date
- Weights
- —
- Speed
- 172 tok/s first token in 0.64 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 Small 3 | 01/30/2025 | 29.2 | #361 | — | — |
| Mistral Small 3.1 | 03/17/2025 | 29.8 | #357 | — | — |
| Mistral Small 3.2 | 06/20/2025 | 26.2 | #383 | — | 1338 |
| Mistral Small 4 | 03/16/2026 | 33.6 | #317 | — | — |
| Mistral Small 4 | 03/16/2026 | 21.6 | #396 | 4.3 | — |
Scores by source
Scores by sourcepercentile = % of ranked models it beats| Metric | Value | Percentile | Source |
|---|---|---|---|
| AA Intelligence Index weight 3 | 4.3 | 2% | Artificial Analysis |
| Humanity's Last Exam weight 1 | 9.9 | 51% | Artificial Analysis |
| Terminal-Bench weight 1 | 0.0 | 22% | Artificial Analysis |
| GPQA Diamond | 76.9 | 62% | Artificial Analysis |
| Long-context reasoning | 49.7 | 44% | Artificial Analysis |
By category
| Category | Score | Rank | Metrics |
|---|---|---|---|
| Agentic | 36.0 | #140 | Terminal-Bench |
| Reasoning | 54.1 | #189 | Humanity's Last Exam, GPQA Diamond |
| Research | 47.1 | #196 | Long-context reasoning |
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.