One-minute worlds · Generated worlds
Still the same place a minute later?
Keep looking away and coming back. See what stays put over a longer visit.
Clip files & individual measurements
Matrix-Game 3 Distilled
Skywork
- AI landmark retention
- 41.7 %
- AI landmark contradiction
- 0.0 %
- AI landmark uncertainty
- 58.3 %
- Return judgments completed
- 100.0 %
- Judge first-attempt coverage
- 100.0 %
- Turn pairs tracked
- 100.0 %
- Rotation pairs qualified
- 100.0 %
- Commanded warp discrepancy
- 0.798 % diagonal
- Qualified rotation error
- 1.052 degrees
- Qualified yaw direction
- 98.9 %
- Idle warp discrepancy
- 0.019 % diagonal
- First decoded worker chunk
- 9.6 seconds
- Steady worker throughput
- 5.087 frames/s
- Peak allocated GPU memory
- 30.994 GiB
- Session process time
- 243.2 seconds
- Late / early worker latency
- 0.972 ratio
Matrix-Game 3 Base
Skywork
- AI landmark retention
- 100.0 %
- AI landmark contradiction
- 0.0 %
- AI landmark uncertainty
- 0.0 %
- Return judgments completed
- 100.0 %
- Judge first-attempt coverage
- 100.0 %
- Turn pairs tracked
- 94.3 %
- Rotation pairs qualified
- 94.3 %
- Commanded warp discrepancy
- 1.053 % diagonal
- Qualified rotation error
- 1.724 degrees
- Qualified yaw direction
- 95.2 %
- Idle warp discrepancy
- 0.018 % diagonal
- First decoded worker chunk
- 166.5 seconds
- Steady worker throughput
- 0.237 frames/s
- Peak allocated GPU memory
- 31.015 GiB
- Session process time
- 4265.1 seconds
- Late / early worker latency
- 0.999 ratio
Each clip shows one registered input and seed. The full study includes every seed and failure.
Look for thisPick a landmark in the opening view. Follow it each time the camera comes back.
Every run, including failures (data file) (opens in a new tab)
Choose a scene or change the models
Examples stay in their registered order. Every seed and any failed runs remain in the full results.
Compare the numbers across every run
Matrix-Game 3 Distilled − Matrix-Game 3 Base. Uses every jointly scored case and seed for these two configurations, across the full study.
| Measurement | Left − right | 95% interval | Paired coverage |
|---|---|---|---|
| AI landmark retentionpercentage points | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| AI landmark contradictionpercentage points | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| AI landmark uncertaintypercentage points | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Return judgments completedpercentage points | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Judge first-attempt coveragepercentage points | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Turn pairs trackedpercentage points | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Rotation pairs qualifiedpercentage points | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Commanded warp discrepancy% diagonal | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Qualified rotation errordegrees | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Qualified yaw directionpercentage points | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Idle warp discrepancy% diagonal | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| First decoded worker chunkseconds | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Steady worker throughputframes/s | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Peak allocated GPU memoryGiB | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Session process timeseconds | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
| Late / early worker latencyratio | Not measured | Interval unavailable | 7 matched pairs4 groups · 1 excluded pairs |
A positive difference means the left model has a higher value; a negative difference means a lower value. Whether that is better depends on the metric. Excluded pairs include inapplicable or unscored measurements. These differences use matched runs and can differ from subtracting the displayed averages. Intervals that cross zero do not establish a clear direction of difference.
Landmark scores are AI estimates, not independently reviewed memory tests. One run ended at our rental deadline; incomplete averages are withheld.
Results corrected on 2026-10-16
Preserve original raw videos and worker snapshots. For SANA score registered camera timestamp f using saved output frame f-1; evaluate interframe motion against camera poses[1:1+n_output]. MG3 retains frame f and poses[0:n_output]. This source-defined alignment correction preceded all primary scoring; original outputs are retained.
Analysis and original results (opens in a new tab)Results corrected on 2026-10-16
Post-registration execution amendment: one judge request failed before returning a label. One separately recorded retry used identical images and configuration. Both attempts, first-attempt coverage and the unknown-cost reservation are preserved; no generation was replaced.
Analysis and original results (opens in a new tab)How we tested it — results, method & every run
Do generated worlds remember the way back?
Which released configuration preserves recognizable scene landmarks across repeated one- and two-minute turn-away/return sessions, follows camera actions, and sustains generation throughput?
4 cases · 2 seeds · 3 compared configurations · completed 2026-10-16
What we found. Four familiar scenes and two seeds test repeated returns over a minute. Landmark scores are AI estimates, not validated game-state memory. Compare MG3 Base and Distilled within their shared next-action interface; SANA receives a future camera path and caption. Worker speed excludes browser delivery. 1 session hit the operator rental deadline; its partial evidence is retained and incomplete balanced averages are withheld. This is not a measured model-quality failure.
Watch the turn-away/return cycles and compare the starting scene with later returns. Scores concern three declared visible landmarks, not every object. SANA has extra future-camera and caption inputs; cross-group clips are illustrative, without paired rankings.
| Model configuration | AI landmark retention% · higher is better | AI landmark contradiction% · lower is better | AI landmark uncertainty% · lower is better | Return judgments completed% · higher is better | Judge first-attempt coverage% · higher is better | Turn pairs tracked% · higher is better | Rotation pairs qualified% · higher is better | Commanded warp discrepancy% diagonal · lower is better | Qualified rotation errordegrees · lower is better | Qualified yaw direction% · higher is better | Idle warp discrepancy% diagonal · lower is better | First decoded worker chunkseconds · lower is better | Steady worker throughputframes/s · higher is better | Peak allocated GPU memoryGiB · lower is better | Session process time |
|---|
