card: final results from the 2026-09-24 run
Browse files
README.md
CHANGED
|
@@ -15,7 +15,7 @@ A 33.6M-parameter network that guesses how many moves a Rubik's Cube is from
|
|
| 15 |
solved, plus a search that uses the guess to find the moves. It was trained from
|
| 16 |
nothing: no solver, no example solutions, no human data.
|
| 17 |
|
| 18 |
-
- Solves **
|
| 19 |
- Solves **every** cube scrambled with 10 moves or fewer
|
| 20 |
- Built as a control for [RubiksBench](https://brianbriscoe.dev/lab/rubiksbench),
|
| 21 |
where language models must write the whole answer in one reply. On the same
|
|
@@ -80,13 +80,22 @@ full scrambles from 16% to 96%.
|
|
| 80 |
Canonical RubiksBench scrambles (the same cubes every language model on the
|
| 81 |
board is tested on), 25 cubes per scramble length:
|
| 82 |
|
| 83 |
-
| Scramble length | Solved (beam 1024) |
|
| 84 |
-
|---|---|
|
| 85 |
-
| 3 |
|
| 86 |
-
| 6 |
|
| 87 |
-
| 10 |
|
| 88 |
-
| 15 |
|
| 89 |
-
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
|
| 91 |
Past about 15 moves, every scramble is fully mixed, so 25, 50 or 100 moves are
|
| 92 |
no harder than 20. A full scramble takes about 80 to 90 seconds on one RTX A6000.
|
|
|
|
| 15 |
solved, plus a search that uses the guess to find the moves. It was trained from
|
| 16 |
nothing: no solver, no example solutions, no human data.
|
| 17 |
|
| 18 |
+
- Solves **95%** of fully scrambled cubes (142 of 150, scrambles of 20 to 100 moves)
|
| 19 |
- Solves **every** cube scrambled with 10 moves or fewer
|
| 20 |
- Built as a control for [RubiksBench](https://brianbriscoe.dev/lab/rubiksbench),
|
| 21 |
where language models must write the whole answer in one reply. On the same
|
|
|
|
| 80 |
Canonical RubiksBench scrambles (the same cubes every language model on the
|
| 81 |
board is tested on), 25 cubes per scramble length:
|
| 82 |
|
| 83 |
+
| Scramble length | Solved (beam 1024) | Average answer |
|
| 84 |
+
|---|---|---|
|
| 85 |
+
| 3 | 25/25 | 3.0 moves |
|
| 86 |
+
| 6 | 25/25 | 6.0 |
|
| 87 |
+
| 10 | 25/25 | 10.0 |
|
| 88 |
+
| 15 | 25/25 | 20.4 |
|
| 89 |
+
| 20 | 24/25 | 22.0 |
|
| 90 |
+
| 25 | 24/25 | 21.9 |
|
| 91 |
+
| 30 | 25/25 | 21.6 |
|
| 92 |
+
| 40 | 23/25 | 23.0 |
|
| 93 |
+
| 50 | 23/25 | 21.9 |
|
| 94 |
+
| 100 | 23/25 | 21.5 |
|
| 95 |
+
| **all** | **242/250 (96.8%)** | |
|
| 96 |
+
|
| 97 |
+
Measured 2026-09-24 on one RTX A6000. Every cube and answer is replayable at
|
| 98 |
+
https://brianbriscoe.dev/lab/tiny-cube-solver.
|
| 99 |
|
| 100 |
Past about 15 moves, every scramble is fully mixed, so 25, 50 or 100 moves are
|
| 101 |
no harder than 20. A full scramble takes about 80 to 90 seconds on one RTX A6000.
|