Up to four models
GPT-5.5 vs DeepSeek-V4-Flash (Apr 2026)
On the index (GPT-4 = 100), GPT-5.5 scores 172 and DeepSeek-V4-Flash (Apr 2026) 142. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
GPT-5.5
172156–188
WeightsClosed
ReleasedApr 2026
Price in / out$5.0 / $30
Context1.05M
Core tests taken6 of 9
142125–159
WeightsOpen
ReleasedApr 2026
Price in / out$0.090 / $0.18
Context1.05M
Core tests taken0 of 9
GPT-5.5 leads DeepSeek-V4-Flash (Apr 2026) by 30 points, at 100× the price.
Test by testthird-party results
GPT-5.5
DeepSeek-V4-Flash (Apr 2026)
Every result as a table
| Test | GPT-5.5 | DeepSeek-V4-Flash (Apr 2026) |
|---|---|---|
| Reasoning & knowledge | ||
| LiveBench Data Analysis | 81.6% | 68.0% |
| LiveBench IF | 70.7% | 63.1% |
| LiveBench Language | 87.4% | 70.1% |
| LiveBench Reasoning | 89.6% | 70.6% |
| SimpleBench | 69.0% | 46.3% |
| Maths | ||
| LiveBench Mathematics | 95.9% | 79.7% |
| Coding | ||
| LiveBench Agentic Coding | 54.0% | 37.6% |
| LiveBench Coding | 82.2% | 69.2% |
| WeirdML | 84.9% | 45.6% |
| Agents | ||
| DTBench | 96.0% | 86.4% |
| LMCA | 54.3% | 35.9% |
Tests at least two of these models have taken. Bold rows are the core tests.