Self-Reference Count
Self Reference
Pass rate
0%
Survived 3 out of 15 breakers
Claude 3 Haiku is Anthropic's fastest and most compact model for near-instant responsiveness. Quick and accurate targeted performance. See the launch announcement and benchmark results [here](https://www.anthropic.com/news/claude-3-haiku) #multimodal
200,000 tokens
$0.25 /1M tokens
$1.25 /1M tokens
4,096
| Test | Category | Latest Result | Success Rate | |
|---|---|---|---|---|
| Self-Reference Count | Self Reference | 0% | ||
| Reverse Word Test | Character Manipulation | 0% | ||
| Alice's Brother Problem | Logic Reasoning | 0% | ||
| Broken Mug | Lateral Thinking | 0% | ||
| Car Wash Dilemma | Logic Reasoning | 0% | ||
| The Missing A | Pattern Matching | 0% | ||
| Horse Race Logic | Logic Reasoning | 0% | ||
| The Compartment Trick | Logic Reasoning | 0% | ||
| Sycophancy Trap | Logic Reasoning | 0% | ||
| Contradictory Premises | Logic Reasoning | 10% | ||
| 10-Step Instructions | Instruction Following | 20% | ||
| Silence Protocol | Instruction Following | 30% | ||
| Strawberry Problem | Character Counting | 100% | ||
| Bullshit Detector | Epistemic Humility | 100% | ||
| Coin Flip Paradox | Logic Reasoning | 100% |