Instruction Following Benchmark as reported in the LLM Benchmark Matrix. Tests a model's ability to precisely follow complex, multi-constraint instructions.
This definition has not completed editorial and provenance review. Verify its metric, version, source, and evaluation configuration upstream before using it for a model-selection decision.
Directly tests instruction-following fidelity — a core capability for assistant models.
Matrix snapshot — we also have AA's own IFBench column.
Higher is better. Accuracy percentage (0–100%).
Matrix snapshot — we also have AA's own IFBench column.
| # | Model | Score |
|---|---|---|
| 1 | 76.5 | |
| 2 | 76.5 | |
| 3 | 75.7 | |
| 4 | 74.1 | |
| 5 | 72.8 | |
| 6 | 72.4 | |
| 7 | 72.3 | |
| 8 | 71.8 | |
| 9 | 71.0 | |
| 10 | 71.0 | |
| 11 | 71.0 | |
| 12 | 70.2 | |
| 13 | 69.6 | |
| 14 | 69.3 | |
| 15 | 69.3 | |
| 16 | 69.2 | |
| 17 | 69.2 | |
| 18 | 69.1 | |
| 19 | 68.9 | |
| 20 | 68.6 | |
| 21 | 68.2 | |
| 22 | 67.9 | |
| 23 | 67.1 | |
| 24 | 66.9 | |
| 25 | 66.9 | |
| 26 | 66.8 | |
| 27 | 66.6 | |
| 28 | 66.6 | |
| 29 | 66.0 | |
| 30 | 65.9 | |
| 31 | 65.6 | |
| 32 | 65.6 | |
| 33 | 65.3 | |
| 34 | 65.3 | |
| 35 | 65.1 | |
| 36 | 65.0 | |
| 37 | 64.9 | |
| 38 | 64.9 | |
| 39 | 64.9 | |
| 40 | 64.4 | |
| 41 | 64.4 | |
| 42 | 64.3 | |
| 43 | 64.2 | |
| 44 | 64.1 | |
| 45 | 64.0 | |
| 46 | 64.0 | |
| 47 | 63.9 | |
| 48 | 63.7 | |
| 49 | 63.4 | |
| 50 | 63.4 |