Anthropic Claude 3 Haiku vs DeepSeek DeepSeek-Math — side-by-side specs.
Anthropic Claude 3 Haiku | DeepSeek DeepSeek-Math | |
|---|---|---|
| Overview | ||
| Company | Anthropic | DeepSeek |
| Release date | Mar 4 2024 | Apr 3 2024 |
| Model type | — | — |
| Open source | No | Yes |
| Specifications | ||
Parameters | — | — |
Context window | — | — |
| Benchmarks | ||
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 33.3% | — |
| Timeline | ||
| Release gap | Claude 3 Haiku shipped 30 days before DeepSeek-Math | |
Claude 3 Haiku and DeepSeek-Math don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Claude 3 Haiku shipped 30 days before DeepSeek-Math, so benchmark comparisons should account for the intervening progress.
DeepSeek-Math is an open-source / open-weight model; Claude 3 Haiku is proprietary.
Direct benchmark comparisons are unavailable — Claude 3 Haiku and DeepSeek-Math don't publish scores on any of the same benchmarks.