Claude 3 HaikuvsClaude Opus 4
Claude 3 Haiku | Claude Opus 4 | |
|---|---|---|
| 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. | 10% | 34%Best |
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. | — | 72.5% |
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% | 79.6%Best |
| Overview | ||
| Company | Anthropic | Anthropic |
| Release date | Mar 4 2024 | May 22 2025 |
| Access | Proprietary | Proprietary |
Which is better: Claude 3 Haiku or Claude Opus 4?
Claude Opus 4 leads Claude 3 Haiku on 2 of the 2 benchmarks they both report (BullshitBench v2, GPQA Diamond). Claude 3 Haiku shipped 444 days before Claude Opus 4, so benchmark comparisons should account for the intervening progress.
Published specifications for these two models are limited — see each model page for the latest details.
On BullshitBench v2, Claude Opus 4 leads at 34% vs Claude 3 Haiku at 10%. On GPQA Diamond, Claude Opus 4 leads at 79.6% vs Claude 3 Haiku at 33.3%.
Frequently asked questions
Claude 3 Haiku was released by Anthropic on Mar 4 2024.
Claude Opus 4 was released by Anthropic on May 22 2025.
Claude Opus 4 leads on GPQA Diamond — Claude 3 Haiku 33.3% vs Claude Opus 4 79.6%.
Other comparisons
Claude 3 HaikuvsGPT-5.6-CyberClaude Opus 4vsGPT-5.6-CyberClaude 3 HaikuvsGemini 3.7 FlashClaude Opus 4vsGemini 3.7 FlashClaude 3 HaikuvsMuse GlimmerClaude Opus 4vsMuse GlimmerClaude 3 HaikuvsGrok 4.6Claude Opus 4vsGrok 4.6Claude 3 HaikuvsDeepSeek-V4-Pro-0813Claude Opus 4vsDeepSeek-V4-Pro-0813Claude 3 HaikuvsMistral Medium 3.5Claude Opus 4vsMistral Medium 3.5