ANLI
Rank #50 of 51 · 9 models scored · top score 37.15
unknown
Strongest on harness comparability, weakest on contamination resistance. Caveat: the public test set leaves contamination on the table.
Weighted composite. Each signal is computed from the source-backed leaderboard below.
Does it still separate models?
How far from ceiling / clustered at the top?
Public vs held-out; training-leak risk.
Apples-to-apples, or mixed / vendor-optimized?
How old is the benchmark?
| # | Model | Score | Evidence | Measured |
|---|---|---|---|---|
| 1 | phi-3-small 7.4B | 37.15 | self-reported· optimized | 2024-04-23 |
| 2 | GPT-3.5 Turbo (Nov 2023) | 37.15 | self-reported· optimized | 2023-06-13 |
| 3 | Llama 3-8B | 35.95 | self-reported· optimized | 2024-04-18 |
| 4 | phi-3-medium 14B | 33.7 | self-reported· optimized | 2024-04-23 |
| 5 | Mixtral 8x7B | 32.8 | self-reported· optimized | 2023-12-11 |
| 6 | phi-3-mini 3.8B | 29.2 | self-reported· optimized | 2024-04-23 |
| 7 | Gemma 7B | 23.05 | self-reported· optimized | 2024-02-21 |
| 8 | Mistral 7B v0.1 | 20.65 | self-reported· optimized | 2023-10-10 |
| 9 | Phi-2 | 13.75 | self-reported· optimized | 2023-12-12 |