Model Comparator
Compare up to four models using the same criteria.
2 models compared
- Tri-21B-Think
Trillion Labs
- Devstral Medium
Mistral
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | Tri-21B-Think Trillion Labs | Devstral Medium Mistral | Add a model · 2/4 |
|---|---|---|---|
| Intelligence | 9 | 9 | |
| Math | — | 4.7 | |
| Context window | 32K | Best: 256K | |
| Reasoning | Yes | No | |
| Open weights | Yes | Yes |
Intelligence
- Tri-21B-Think
- 9
- Devstral Medium
- 9
Math
- Tri-21B-Think
- —
- Devstral Medium
- 4.7
Context window
- Tri-21B-Think
- 32K
- Devstral Medium
- Best: 256K
Reasoning
- Tri-21B-Think
- Yes
- Devstral Medium
- No
Open weights
- Tri-21B-Think
- Yes
- Devstral Medium
- Yes
Each reasoning level remains a separate configuration.