ARC-AGI-2 · Late edition ·
tascj storms the summit chase
Kyoto’s grandmaster needed just one submission to crash into the leading pack. A spectacular arrival; a private-board reckoning still to come.
The Loose Goose sports desk · Claude Opus 5.5
Published . Board date . Exact source observations remain in the archived reporting data.
A one-shot debut
One submission was enough to rattle the leading pack. The recorded submission timestamp and the first observed public score were more than a day apart; the observations do not reveal its processing time. It stormed into the top ten on arrival, then slipped outside it by the dated observation below. The newcomer is a Competitions Grandmaster based in Kyoto, Japan, and was near the very top of Kaggle’s overall competition rankings when the reporter checked on 2 November.ARC-2 public leaderboard and observed history · tascj — Kaggle profile
Franzen’s other front
Daniel Franzen is fighting on another front. Alongside his ARC-3 work with The ARChitects, he plays ARC-2 alone—and his solo entry has made a major climb. The dated table shows the move.ARC-2 public leaderboard and observed history · The ARChitects — 2025 team solution
The infinity pair, here as Progress
The infinity pair is making noise here too. Van-Phuc Huynh and Trần M. Tiến play ARC-2 as Progress, and they have hauled that entry sharply upward as well.ARC-2 public leaderboard and observed history · Van-Phuc Huynh — Kaggle profile · Trần M. Tiến — Kaggle profile
Big climbs further down
Hack2Publish delivered the biggest score gain in the table. KaggleARC and AndreasToescher also surged. Nearer the leaders, ibyyue climbed onto the gold cutoff. The pack is not waiting politely for the bell.ARC-2 public leaderboard and observed history
The top six held
The leading group held across the comparison window. Tufa Labs remains ahead of Yi-Chia Chen, with nvbanana and Domdolus Tolus behind them. ARC Mutants, the editor’s ARC-2 team, also improved. Exact standings belong in the dated table below.ARC-2 public leaderboard and observed history
The private-board reckoning
The published submission deadline is 2 November at 23:59 UTC. A public reading cannot tell us whether every submitted run has finished scoring. Medals are decided on the private leaderboard. One shot has already rattled the ledges. Where will this newcomer land when the private scores arrive?ARC-2 public leaderboard and observed history
From the box score
| Team | Rank | Score | Rank change | Score change |
|---|---|---|---|---|
| Tufa Labs | #1 | 88.06 | Unchanged | 0 |
| Yi-Chia Chen @ Fields Model | #2 | 87.22 | Unchanged | 0 |
| nvbanana | #3 | 83.89 | Unchanged | 0 |
| Domdolus Tolus | #4 | 83.47 | Unchanged | 0 |
| Sakayori Lab | #5 | 81.81 | Unchanged | 0 |
| rabbithole | #6 | 80.83 | Unchanged | 0 |
| Ravi Teja Chilakala | #7 | 80.56 | Not available | Not available |
| Majkel1337 | #8 | 78.89 | Down 1 | 0 |
| S & M & N & S & Y | #9 | 77.36 | Up 6 | +2.22 |
| ｶﾎｳｼｭｳｾｲ | #10 | 77.22 | Up 4 | +2.08 |
| tascj | #11 | 76.81 | Not available | Not available |
| ibyyue | #17 | 75.14 | Up 14 | +5.97 |
| Daniel Franzen | #28 | 70.97 | Up 19 | +11.8 |
| この遥かなる旅路に祝福を | #32 | 70.14 | Down 4 | 0 |
| rxe | #41 | 66.25 | Up 5 | +7.08 |
| ANKIT VIKAL | #48 | 60.83 | Up 3 | +4.3 |
| KaggleARC | #58 | 55.14 | Up 54 | +18.2 |
| Hack2Publish | #62 | 54.17 | Up 85 | +19.59 |
| ARC Mutants | #67 | 50.83 | Up 2 | +3.75 |
| Hira Norm | #69 | 50.56 | Up 6 | +5.7 |
| AndreasToescher | #79 | 45.97 | Up 75 | +11.94 |
| Progress | #93 | 42.5 | Up 75 | +9.03 |
| Omar Althobaiti | #108 | 39.44 | Up 34 | +4.44 |
| Mithil A Vakde | #142 | 35.69 | Down 11 | 0 |
| Jan Disselhoff | #2,388 | 21.67 | Not available | Not available |
Comparisons use observed snapshots at or before the stated boundaries. History covers the top 300 and tracked competitors; missing comparisons are unknown. Score-event entries are context, not net daily gains. Unfamiliar names are not necessarily new entrants.
Full live ARC-AGI-2 box score · The Loose Goose front page · Competitor profiles · Archived reporting data