Self-Reference Count
Self Reference
Pass rate
0%
Survived 5 out of 15 breakers
Mistral Large 3 2512 is Mistral’s most capable model to date, featuring a sparse mixture-of-experts architecture with 41B active parameters (675B total), and released under the Apache 2.0 license.
262,144 tokens
$0.50 /1M tokens
$1.50 /1M tokens
–
| Test | Category | Latest Result | Success Rate | |
|---|---|---|---|---|
| Self-Reference Count | Self Reference | 0% | ||
| 10-Step Instructions | Instruction Following | 0% | ||
| Silence Protocol | Instruction Following | 0% | ||
| Contradictory Premises | Logic Reasoning | 0% | ||
| Car Wash Dilemma | Logic Reasoning | 0% | ||
| The Missing A | Pattern Matching | 0% | ||
| Coin Flip Paradox | Logic Reasoning | 0% | ||
| Bullshit Detector | Epistemic Humility | 25% | ||
| Horse Race Logic | Logic Reasoning | 25% | ||
| The Compartment Trick | Logic Reasoning | 25% | ||
| Broken Mug | Lateral Thinking | 50% | ||
| Strawberry Problem | Character Counting | 100% | ||
| Reverse Word Test | Character Manipulation | 100% | ||
| Alice's Brother Problem | Logic Reasoning | 100% | ||
| Sycophancy Trap | Logic Reasoning | 100% |