Claude Haiku 4.5 vs DeepSeek V2.5 (Revised)
Anthropic Claude Haiku 4.5 | DeepSeek DeepSeek V2.5 (Revised) | |
|---|---|---|
| Overview | ||
| Company | Anthropic | DeepSeek |
| Release date | Oct 15 2025 | Dec 2024 |
| Access | Proprietary | Open Weight |
| 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. | 77% | — |
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. | 73.3% | — |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 73% | — |
| Timeline | ||
| Release gap | DeepSeek V2.5 (Revised) shipped 318 days before Claude Haiku 4.5 | |
Which is better: Claude Haiku 4.5 or DeepSeek V2.5 (Revised)?
Claude Haiku 4.5 and DeepSeek V2.5 (Revised) don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. DeepSeek V2.5 (Revised) shipped 318 days before Claude Haiku 4.5, so benchmark comparisons should account for the intervening progress.
Claude Haiku 4.5 is proprietary, while DeepSeek V2.5 (Revised) is open weight.
Direct benchmark comparisons are unavailable — Claude Haiku 4.5 and DeepSeek V2.5 (Revised) don't publish scores on any of the same benchmarks.
Frequently asked questions
Claude Haiku 4.5 was released by Anthropic on Oct 15 2025.
DeepSeek V2.5 (Revised) was released by DeepSeek on Dec 2024.
Claude Haiku 4.5 is a proprietary model released by Anthropic. DeepSeek V2.5 (Revised) is an open weight model released by DeepSeek.