Claude 3.5 HaikuvsClaude 3.5 Sonnet
Claude 3.5 Haiku | Claude 3.5 Sonnet | |
|---|---|---|
| Benchmarks | ||
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. | 50% | 46% |
SWE-Bench VerifiedCoding — Real 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% | 33.4% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 41.6% | 59.4% |
| Overview | ||
| Company | Anthropic | Anthropic |
| Release date | Oct 22 2024 | Jun 20 2024 |
| Access | Proprietary | Proprietary |
Other comparisons
Claude 3.5 HaikuvsGPT-6 AstraClaude 3.5 SonnetvsGPT-6 AstraClaude 3.5 HaikuvsGemini 3.8 FlashClaude 3.5 SonnetvsGemini 3.8 FlashClaude 3.5 HaikuvsMuse Spark 1.3Claude 3.5 SonnetvsMuse Spark 1.3Claude 3.5 HaikuvsGrok 4.6Claude 3.5 SonnetvsGrok 4.6Claude 3.5 HaikuvsDeepSeek-V4.1-FlashClaude 3.5 SonnetvsDeepSeek-V4.1-FlashClaude 3.5 HaikuvsMistral Medium 3.5Claude 3.5 SonnetvsMistral Medium 3.5
Frequently asked questions
Claude 3.5 Haiku leads Claude 3.5 Sonnet on 2 of the 3 benchmarks they both report (BullshitBench v2, SWE-Bench Verified, GPQA Diamond). Claude 3.5 Sonnet shipped 124 days before Claude 3.5 Haiku, 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 3.5 Haiku leads at 50% vs Claude 3.5 Sonnet at 46%. On SWE-Bench Verified, Claude 3.5 Haiku leads at 40.6% vs Claude 3.5 Sonnet at 33.4%. On GPQA Diamond, Claude 3.5 Sonnet leads at 59.4% vs Claude 3.5 Haiku at 41.6%.