Claude Opus 4.7 vs DeepSeek-V3.1
Anthropic vs DeepSeek — benchmarks, pricing, and capabilities side by side.
- •Claude Opus 4.7 has the higher intelligence index (90.9 vs 59.8)
- •DeepSeek-V3.1 is cheaper ($0.21 vs $5.00 per 1M input)
- •Claude Opus 4.7 has a larger context window (1M)
| Claude Opus 4.7 | DeepSeek-V3.1 | |
|---|---|---|
| Intelligence index | 90.9 | 59.8 |
| Developer | Anthropic | DeepSeek |
| Type | LLM | LLM |
| Access | API only | Open weights |
| Context window | 1,000,000 tokens | 163,840 tokens |
| Input price | $5.00 / 1M | $0.21 / 1M |
| Output price | $25.00 / 1M | $0.79 / 1M |
| Speed | 49 tok/s | — |
| Released | April 16, 2026 | August 21, 2025 |
| Parameters | — | 671B (37B active) |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
Shared benchmarks
Claude Opus 4.7
DeepSeek-V3.1
GPQA Diamond
94.2
74.9
Humanity’s Last Exam
39.6
15.9
SWE-bench Verified
87.6
66