DeepSeek-V4-Provso3
DeepSeek-V4-Pro | o3 | |
|---|---|---|
| Benchmarks | ||
Nonsense detection BullshitBench v2Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 14% | 26%Best |
Competitive coding LiveCodeBenchCoding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | 93.5% | — |
Web browsing BrowseCompCan the AI browse the web and track down hard-to-find answers? Higher is better. | 83.4% | — |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 90.1% | — |
Community preference (code) Arena Elo (Code)Like the text arena, but people vote on which AI writes better code. The votes become a chess-style Elo rating on arena.ai. Higher is better. | 1457 | — |
| Timeline | ||
| Release gap | o3 shipped 373 days before DeepSeek-V4-Pro | |
| Overview | ||
| Company | DeepSeek | OpenAI |
| Release date | Apr 24 2026 | Apr 16 2025 |
| Access | Open Weight | Proprietary |
Which is better: DeepSeek-V4-Pro or o3?
o3 leads DeepSeek-V4-Pro on 1 of the 1 benchmark they both report (BullshitBench v2). o3 shipped 373 days before DeepSeek-V4-Pro, so benchmark comparisons should account for the intervening progress.
DeepSeek-V4-Pro is open weight, while o3 is proprietary.
On BullshitBench v2, o3 leads at 26% vs DeepSeek-V4-Pro at 14%.
Frequently asked questions
DeepSeek-V4-Pro was released by DeepSeek on Apr 24 2026.
o3 was released by OpenAI on Apr 16 2025.
DeepSeek-V4-Pro is an open weight model released by DeepSeek. o3 is a proprietary model released by OpenAI.