fedi-index:數學
數學推理:Arena 數學投票與 Epoch AI 的 FrontierMath 測試結果。
| # | 等級 | 模型 | fedi-index | 百分位 | 人類偏好 | 解題能力 | 性價比 | 來源 | 分數拆解 |
|---|---|---|---|---|---|---|---|---|---|
| 19 | A | 1,456 ±9 | 93 | 1,462 | 1,449 | +44 | AE | › | |
| 25 | A | 1,448 ±10 | 91 | 1,457 | 1,440 | +60 | AE | › | |
| 28 | A | 1,443 ±9 | 89 | 1,468 | 1,416 | +105◆ | AE | › | |
| 29 | A | 1,442 ±8 | 89 | 1,455 | 1,426 | +45 | AE | › | |
| 30 | A | 1,438 ±9 | 89 | 1,454 | — | +80 | A | › | |
| 33 | A | 1,436 ±8 | 87 | 1,447 | 1,421 | +44 | AE | › | |
| 38 | A | 1,430 ±8 | 85 | 1,444 | — | +36 | A | › | |
| 42 | A | 1,426 ±21 | 84 | 1,449 | — | +92◆ | A | › | |
| 45 | A | 1,418 ±20 | 83 | 1,439 | — | +68 | A | › | |
| 46 | A | 1,418 ±8 | 82 | 1,446 | 1,385 | +36 | AE | › | |
| 49 | A | 1,416 ±8 | 81 | 1,449 | 1,381 | +27 | AE | › | |
| 51 | A | 1,415 ±24 | 80 | 1,439 | — | +98◆ | A | › | |
| 52 | B | 1,415 ±9 | 80 | 1,426 | 1,403 | +38 | AE | › | |
| 53 | B | 1,415 ±9 | 80 | — | 1,433 | +40 | E | › | |
| 54 | B | 1,413 ±7 | 79 | 1,430 | 1,395 | +33 | AE | › | |
| 55 | B | 1,413 ±16 | 79 | 1,430 | — | +46 | A | › | |
| 56 | B | 1,412 ±22 | 78 | 1,433 | — | +54 | A | › | |
| 60 | B | 1,409 ±21 | 77 | 1,430 | — | +59 | A | › | |
| 65 | B | 1,405 ±19 | 75 | 1,424 | — | +33 | A | › | |
| 66 | B | 1,405 ±7 | 74 | 1,430 | 1,374 | +27 | AE | › | |
| 68 | B | 1,404 ±9 | 74 | — | 1,422 | +89 | E | › | |
| 70 | B | 1,401 ±12 | 73 | 1,415 | — | +23 | A | › | |
| 78 | B | 1,399 ±10 | 70 | 1,412 | — | +75 | A | › | |
| 79 | B | 1,399 ±14 | 69 | 1,414 | — | +32 | A | › | |
| 82 | B | 1,396 ±8 | 68 | — | 1,412 | +9 | E | › | |
| 84 | B | 1,395 ±11 | 67 | 1,408 | — | +26 | A | › | |
| 85 | B | 1,395 ±10 | 67 | 1,408 | — | +50 | A | › | |
| 86 | B | 1,395 ±11 | 66 | 1,408 | — | +14 | A | › | |
| 91 | B | 1,392 ±12 | 65 | 1,405 | — | +14 | A | › | |
| 94 | B | 1,392 ±10 | 63 | 1,404 | — | +71 | A | › | |
| 97 | B | 1,390 ±13 | 62 | 1,403 | — | +18 | A | › | |
| 100 | B | 1,388 ±8 | 61 | 1,400 | — | +32 | A | › | |
| 101 | B | 1,387 ±19 | 61 | 1,404 | — | -19 | A | › | |
| 103 | B | 1,383 ±17 | 60 | 1,398 | — | +20 | A | › | |
| 104 | B | 1,383 ±18 | 59 | 1,399 | — | +27 | A | › | |
| 106 | B | 1,382 ±15 | 59 | 1,396 | — | +19 | A | › | |
| 110 | B | 1,381 ±16 | 57 | 1,395 | — | +46 | A | › | |
| 116 | B | 1,377 ±9 | 55 | 1,388 | — | +50 | A | › | |
| 119 | B | 1,371 ±11 | 54 | 1,383 | — | +11 | A | › | |
| 121 | B | 1,369 ±10 | 53 | — | 1,382 | +2 | E | › | |
| 123 | B | 1,366 ±9 | 52 | 1,377 | — | +55◆ | A | › | |
| 124 | B | 1,366 ±22 | 52 | 1,382 | — | -17 | A | › | |
| 126 | B | 1,364 ±12 | 51 | 1,376 | — | -2 | A | › | |
| 127 | B | 1,363 ±16 | 50 | 1,376 | — | -10 | A | › | |
| 128 | B | 1,362 ±15 | 50 | 1,375 | — | +18 | A | › | |
| 129 | C | 1,362 ±13 | 50 | 1,374 | — | +25 | A | › | |
| 132 | C | 1,361 ±17 | 48 | 1,374 | — | -22 | A | › | |
| 134 | C | 1,359 ±12 | 48 | 1,370 | — | -2 | A | › | |
| 137 | C | 1,351 ±10 | 46 | 1,361 | — | +9 | A | › | |
| 141 | C | 1,346 ±19 | 45 | 1,359 | — | +0 | A | › | |
| 142 | C | 1,342 ±9 | 44 | — | 1,353 | +2 | E | › | |
| 143 | C | 1,341 ±15 | 44 | 1,351 | — | +6 | A | › | |
| 145 | C | 1,340 ±12 | 43 | 1,350 | — | -30 | A | › | |
| 148 | C | 1,337 ±12 | 42 | 1,347 | — | -22 | A | › | |
| 151 | C | 1,335 ±11 | 41 | 1,344 | — | -36 | A | › | |
| 155 | C | 1,330 ±11 | 39 | 1,339 | — | -29 | A | › | |
| 159 | C | 1,327 ±17 | 38 | 1,337 | — | +1 | A | › | |
| 161 | C | 1,318 ±14 | 37 | 1,327 | — | -14 | A | › | |
| 164 | C | 1,315 ±25 | 36 | 1,328 | — | -32 | A | › | |
| 167 | C | 1,296 ±18 | 35 | 1,304 | — | -26 | A | › | |
| 177 | C | 1,260 ±14 | 31 | 1,264 | — | -95 | A | › | |
| 181 | C | 1,244 ±12 | 29 | 1,247 | — | — | A | › | |
| 196 | D | 1,192 ±16 | 23 | 1,191 | — | -133 | A | › | |
| 198 | D | 1,188 ±8 | 22 | 1,187 | — | -236 | A | › |
AArena(人類投票)EEpoch AI(基準測試) ◆ 帕累托前緣:沒有模型能同時更好又更便宜
fedi-index 是固定尺度上的 Elo 等值:在同一方法論版本內,各月份可互相比較。 等級 S–D:依分項內的百分位決定,且不會高於信賴區間所允許的等級(門檻見方法論)。 「暫定」——此處只有一個來源系列測量過該模型。 方法論