Model Comparator
Compare up to four models using the same criteria.
2 models compared
- DeepSeek V3 0324
DeepSeek
- Tri-21B-think Preview
Trillion Labs
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | DeepSeek V3 0324 DeepSeek | Tri-21B-think Preview Trillion Labs | Add a model · 2/4 |
|---|---|---|---|
| Intelligence | Best: 9.7 | 9.6 | |
| Coding | 21.2 | — | |
| Math | 41 | — | |
| Agentic | 0.8 | — | |
| Speed | 109 t/s | — | |
| TTFT | 0.71 s | — | |
| Input · $ / 1M tokens | $0.24 | — | |
| Output · $ / 1M tokens | $0.90 | — | |
| Context window | Best: 128K | 32K | |
| Reasoning | No | Yes | |
| Open weights | Yes | Yes | |
| Cost / task | $0.0211 | — |
Intelligence
- DeepSeek V3 0324
- Best: 9.7
- Tri-21B-think Preview
- 9.6
Coding
- DeepSeek V3 0324
- 21.2
- Tri-21B-think Preview
- —
Math
- DeepSeek V3 0324
- 41
- Tri-21B-think Preview
- —
Agentic
- DeepSeek V3 0324
- 0.8
- Tri-21B-think Preview
- —
Speed
- DeepSeek V3 0324
- 109 t/s
- Tri-21B-think Preview
- —
TTFT
- DeepSeek V3 0324
- 0.71 s
- Tri-21B-think Preview
- —
Input · $ / 1M tokens
- DeepSeek V3 0324
- $0.24
- Tri-21B-think Preview
- —
Output · $ / 1M tokens
- DeepSeek V3 0324
- $0.90
- Tri-21B-think Preview
- —
Context window
- DeepSeek V3 0324
- Best: 128K
- Tri-21B-think Preview
- 32K
Reasoning
- DeepSeek V3 0324
- No
- Tri-21B-think Preview
- Yes
Open weights
- DeepSeek V3 0324
- Yes
- Tri-21B-think Preview
- Yes
Cost / task
- DeepSeek V3 0324
- $0.0211
- Tri-21B-think Preview
- —
Each reasoning level remains a separate configuration.
Score and cost
Each line joins the reasoning levels of a compared model; the large point marks the configuration in the table. The top-right corner holds the best score for the lowest cost. Select another point to compare that level.
- DeepSeek V3 03249.7 · $0.0211
Not plotted for lack of data on these axes: Tri-21B-think Preview.