Up to four models
Claude Opus 5 vs DeepSeek-V4-Flash
On the index (GPT-4 = 100), Claude Opus 5 scores 185 and DeepSeek-V4-Flash 164. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude Opus 5
185169–201
WeightsClosed
ReleasedJul 2026
Price in / out$5.0 / $25
Context1M
Core tests taken6 of 9
Claude Opus 5 and DeepSeek-V4-Flash are too close to call (21 points apart, within the uncertainty).
Test by testthird-party results
Claude Opus 5
DeepSeek-V4-Flash
Every result as a table
| Test | Claude Opus 5 | DeepSeek-V4-Flash |
|---|---|---|
| Reasoning & knowledge | ||
| LiveBench Data Analysis | 74.6% | 79.5% |
| LiveBench IF | 63.8% | 71.0% |
| LiveBench Language | 88.7% | 80.4% |
| LiveBench Reasoning | 91.2% | 85.4% |
| Maths | ||
| LiveBench Mathematics | 95.7% | 87.8% |
| Coding | ||
| LiveBench Agentic Coding | 65.2% | 65.1% |
| LiveBench Coding | 81.5% | 68.2% |
Tests at least two of these models have taken. Bold rows are the core tests.