Model Comparator
Compare up to four models using the same criteria.
2 models compared
- DeepSeek-V2.5
DeepSeek
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | DeepSeek-V2.5 DeepSeek | Add a model · 2/4 | |
|---|---|---|---|
| Intelligence | 6.6 | 6.6 | |
| Math | — | 4 | |
| Speed | — | 36.2 t/s | |
| TTFT | — | 2.04 s | |
| Input · $ / 1M tokens | — | $0.56 | |
| Output · $ / 1M tokens | — | $0.56 | |
| Context window | 128K | 128K | |
| Reasoning | No | No | |
| Open weights | Yes | Yes |
Intelligence
- DeepSeek-V2.5
- 6.6
- Llama 3.1 Instruct 70B
- 6.6
Math
- DeepSeek-V2.5
- —
- Llama 3.1 Instruct 70B
- 4
Speed
- DeepSeek-V2.5
- —
- Llama 3.1 Instruct 70B
- 36.2 t/s
TTFT
- DeepSeek-V2.5
- —
- Llama 3.1 Instruct 70B
- 2.04 s
Input · $ / 1M tokens
- DeepSeek-V2.5
- —
- Llama 3.1 Instruct 70B
- $0.56
Output · $ / 1M tokens
- DeepSeek-V2.5
- —
- Llama 3.1 Instruct 70B
- $0.56
Context window
- DeepSeek-V2.5
- 128K
- Llama 3.1 Instruct 70B
- 128K
Reasoning
- DeepSeek-V2.5
- No
- Llama 3.1 Instruct 70B
- No
Open weights
- DeepSeek-V2.5
- Yes
- Llama 3.1 Instruct 70B
- Yes
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.
- Llama 3.1 Instruct 70B6.6 · $0.56
Not plotted for lack of data on these axes: DeepSeek-V2.5.