Claude 3.5 HaikuvsDeepSeek R1-0528
Claude 3.5 Haiku | DeepSeek R1-0528 | |
|---|---|---|
| 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. | 50%Best | 8% |
Coding SWE-Bench VerifiedReal coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 40.6% | — |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 41.6% | — |
| Overview | ||
| Company | Anthropic | DeepSeek |
| Release date | Oct 22 2024 | May 28 2025 |
| Access | Proprietary | Open Weight |
Which is better: Claude 3.5 Haiku or DeepSeek R1-0528?
Claude 3.5 Haiku leads DeepSeek R1-0528 on 1 of the 1 benchmark they both report (BullshitBench v2). Claude 3.5 Haiku shipped 218 days before DeepSeek R1-0528, so benchmark comparisons should account for the intervening progress.
Claude 3.5 Haiku is proprietary, while DeepSeek R1-0528 is open weight.
On BullshitBench v2, Claude 3.5 Haiku leads at 50% vs DeepSeek R1-0528 at 8%.
Frequently asked questions
Claude 3.5 Haiku was released by Anthropic on Oct 22 2024.
DeepSeek R1-0528 was released by DeepSeek on May 28 2025.
Claude 3.5 Haiku is a proprietary model released by Anthropic. DeepSeek R1-0528 is an open weight model released by DeepSeek.
Other comparisons
Claude 3.5 HaikuvsGPT-5.6 SolDeepSeek R1-0528vsGPT-5.6 SolClaude 3.5 HaikuvsGemini 3.6 FlashDeepSeek R1-0528vsGemini 3.6 FlashClaude 3.5 HaikuvsMuse Spark 1.2DeepSeek R1-0528vsMuse Spark 1.2Claude 3.5 HaikuvsGrok 4.5DeepSeek R1-0528vsGrok 4.5Claude 3.5 HaikuvsMistral Medium 3.5DeepSeek R1-0528vsMistral Medium 3.5Claude 3.5 HaikuvsKimi K3DeepSeek R1-0528vsKimi K3