Claude 3 HaikuvsClaude 3 Sonnet
Claude 3 Haiku | Claude 3 Sonnet | |
|---|---|---|
| Benchmarks | ||
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 33.3% | 40.4% |
| BenchmarksPublished by one model only | ||
BullshitBench v2Nonsense detection — Given 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% | — |
| Overview | ||
| Company | Anthropic | Anthropic |
| Release date | Mar 4 2024 | Mar 4 2024 |
| Access | Proprietary | Proprietary |
Frequently asked questions
Claude 3 Sonnet leads Claude 3 Haiku on 1 of the 1 benchmark they both report (GPQA Diamond). Both models were released on the same day — Mar 4 2024.
Published specifications for these two models are limited — see each model page for the latest details.
On GPQA Diamond, Claude 3 Sonnet leads at 40.4% vs Claude 3 Haiku at 33.3%.
Claude 3 Haiku was released by Anthropic on Mar 4 2024.
Claude 3 Sonnet was released by Anthropic on Mar 4 2024.
Claude 3 Sonnet leads on GPQA Diamond — Claude 3 Haiku 33.3% vs Claude 3 Sonnet 40.4%.
Other comparisons
Claude 3 HaikuvsGPT-5.6-CyberClaude 3 SonnetvsGPT-5.6-CyberClaude 3 HaikuvsGemini 3.7 FlashClaude 3 SonnetvsGemini 3.7 FlashClaude 3 HaikuvsMuse GlimmerClaude 3 SonnetvsMuse GlimmerClaude 3 HaikuvsGrok 4.6Claude 3 SonnetvsGrok 4.6Claude 3 HaikuvsDeepSeek-V4-Pro-0813Claude 3 SonnetvsDeepSeek-V4-Pro-0813Claude 3 HaikuvsMistral Medium 3.5Claude 3 SonnetvsMistral Medium 3.5