Model Comparator
Compare up to four models using the same criteria.
2 models compared
- DeepSeek V4 Pro 0813 (Max)
DeepSeek
✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.
| Metrics | DeepSeek V4 Pro 0813 (Max) DeepSeek | Add a model · 2/4 | |
|---|---|---|---|
| Intelligence | Best: 23.2 | 23.1 | |
| Coding | 66.4 | Best: 68.8 | |
| Agentic | — | 41.3 | |
| Speed | 59.4 t/s | Best: 72.7 t/s | |
| TTFT | Best: 0.87 s | 1.62 s | |
| Input · $ / 1M tokens | $2×1.5 relative to the lowest price among the compared models | Best: $1.32 | |
| Output · $ / 1M tokens | $10×2.5 relative to the lowest price among the compared models | Best: $3.96 | |
| Context window | 1M | 1M | |
| Reasoning | No | Yes | |
| Open weights | No | Yes | |
| Cost / task | — | $1.77 |
Intelligence
- Claude Sonnet 5 (Non-reasoning, High)
- Best: 23.2
- DeepSeek V4 Pro 0813 (Max)
- 23.1
Coding
- Claude Sonnet 5 (Non-reasoning, High)
- 66.4
- DeepSeek V4 Pro 0813 (Max)
- Best: 68.8
Agentic
- Claude Sonnet 5 (Non-reasoning, High)
- —
- DeepSeek V4 Pro 0813 (Max)
- 41.3
Speed
- Claude Sonnet 5 (Non-reasoning, High)
- 59.4 t/s
- DeepSeek V4 Pro 0813 (Max)
- Best: 72.7 t/s
TTFT
- Claude Sonnet 5 (Non-reasoning, High)
- Best: 0.87 s
- DeepSeek V4 Pro 0813 (Max)
- 1.62 s
Input · $ / 1M tokens
- Claude Sonnet 5 (Non-reasoning, High)
- $2×1.5 relative to the lowest price among the compared models
- DeepSeek V4 Pro 0813 (Max)
- Best: $1.32
Output · $ / 1M tokens
- Claude Sonnet 5 (Non-reasoning, High)
- $10×2.5 relative to the lowest price among the compared models
- DeepSeek V4 Pro 0813 (Max)
- Best: $3.96
Context window
- Claude Sonnet 5 (Non-reasoning, High)
- 1M
- DeepSeek V4 Pro 0813 (Max)
- 1M
Reasoning
- Claude Sonnet 5 (Non-reasoning, High)
- No
- DeepSeek V4 Pro 0813 (Max)
- Yes
Open weights
- Claude Sonnet 5 (Non-reasoning, High)
- No
- DeepSeek V4 Pro 0813 (Max)
- Yes
Cost / task
- Claude Sonnet 5 (Non-reasoning, High)
- —
- DeepSeek V4 Pro 0813 (Max)
- $1.77
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.
- Claude Sonnet 5 (Non-reasoning, High)23.2 · $4
- DeepSeek V4 Pro 0813 (Max)23.1 · $1.98