Model Comparator
Compare up to four models using the same criteria.
2 models compared
- DeepSeek V3 (Dec '24)
DeepSeek
- Devstral 2
Mistral
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | DeepSeek V3 (Dec '24) DeepSeek | Devstral 2 Mistral | Add a model · 2/4 |
|---|---|---|---|
| Intelligence | 8.5 | Best: 8.6 | |
| Coding | 23 | Best: 31.3 | |
| Math | 26 | Best: 36.7 | |
| Agentic | 0.8 | Best: 2.3 | |
| Speed | 37 t/s | Best: 139 t/s | |
| TTFT | Best: 1.4 s | 2.34 s | |
| Input · $ / 1M tokens | $0.32 | — | |
| Output · $ / 1M tokens | $0.89 | — | |
| Context window | 128K | Best: 256K | |
| Reasoning | No | No | |
| Open weights | Yes | Yes | |
| Cost / task | $0.0198 | — |
Intelligence
- DeepSeek V3 (Dec '24)
- 8.5
- Devstral 2
- Best: 8.6
Coding
- DeepSeek V3 (Dec '24)
- 23
- Devstral 2
- Best: 31.3
Math
- DeepSeek V3 (Dec '24)
- 26
- Devstral 2
- Best: 36.7
Agentic
- DeepSeek V3 (Dec '24)
- 0.8
- Devstral 2
- Best: 2.3
Speed
- DeepSeek V3 (Dec '24)
- 37 t/s
- Devstral 2
- Best: 139 t/s
TTFT
- DeepSeek V3 (Dec '24)
- Best: 1.4 s
- Devstral 2
- 2.34 s
Input · $ / 1M tokens
- DeepSeek V3 (Dec '24)
- $0.32
- Devstral 2
- —
Output · $ / 1M tokens
- DeepSeek V3 (Dec '24)
- $0.89
- Devstral 2
- —
Context window
- DeepSeek V3 (Dec '24)
- 128K
- Devstral 2
- Best: 256K
Reasoning
- DeepSeek V3 (Dec '24)
- No
- Devstral 2
- No
Open weights
- DeepSeek V3 (Dec '24)
- Yes
- Devstral 2
- Yes
Cost / task
- DeepSeek V3 (Dec '24)
- $0.0198
- Devstral 2
- —
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 (Dec '24)8.5 · $25.37
- Devstral 28.6 · $0