Beam Free
Key numbers
- Panel score
- 3not ranked, failed run
- Cost
- $0free model
- Wall time
- 8.0 min
- Model time
- 8.0 min100% of wall time
- tok/s
- 12.3
- Tokens
- 224kOutput: 1.4k
- Cache hit
- 86%
- API calls
- 26
- Tool calls
- 9
- Renderer
- Intel UHD 630 (GPU)
- H1–H6
- Pass
- Cost per point
- $0.000
How the run ended
Knowingly delivered a placeholder: random boxes on a dark plane, a line for the route and a box for the drone. The file passes the format checks H1–H6, but there are no landmarks, no people and no contact sheet.
Blind panel
Scores of the three judges in every panel this video was part of.
| Panel | Letter | Anthropic | OpenAI | Median | Spread | Rank | |
|---|---|---|---|---|---|---|---|
| 14 Nov 2026 11 videos | I | 3 | 3 | 4 | 3 | 1 | 11 / 11 |
| 13 Nov 2026 10 videos | E | 5 | 4 | 7 | 5 | 3 | 10 / 10 |
| 11 Nov 2026 9 videos | H | 2 | 3 | 6 | 3 | 4 | 9 / 9 |
Sub-scores and the judges’ one-line notes from the panel of 14 Nov 2026.
| judge | Landmarks / 25 | City look / 25 | Camera & drone / 30 | Life / 10 | Overall / 10 | Total |
|---|---|---|---|---|---|---|
| Anthropic | 0 | 1 | 2 | 0 | 0 | 3 |
| OpenAI | 0 | 1 | 2 | 0 | 0 | 3 |
| 2 | 1 | 1 | 0 | 0 | 4 |
Judges’ notes
- AnthropicBroken render: static flat plane with scattered boxes and no recognisable city, landmarks or life.
- OpenAIA dark plane, scattered boxes and an indistinct foreground object provide almost none of the requested scene.
- GoogleBroken rendering across all 16 frames with the lower two-thirds pitch black and only a thin distant horizon strip visible below the sky.
Time and speed
Wall time includes rendering and encoding. Model time is the sum of all model calls.
- Model generating
- Tools & waiting
| Wall time | 8.0 min |
|---|---|
| Model time | 8.0 min |
| Tools & waiting | 0.0 min |
| Model share of wall time | 100% |
| tok/s | 12.3 |
| Model calls | 26 |
Tokens and cost
Tokens summed over every model response. Tokens multiplied by the provider’s price.
| Uncached input | 31,036 |
|---|---|
| Cache read | 191,168 |
| Cache write | 0 |
| Output | 1,391of which 534 reasoning |
| Total | 223,595 |
| Cache-hit ratio | 86.0% |
| Cost | $0free model |
| Output tokens per point | 464 |
Tools used
Tool calls by name, as the harness logged them. Names differ between harnesses.
Rendering
| Renderer | Intel UHD 630 (GPU) |
|---|---|
| Final render | 10 min |
| Per frame | – |
Hard requirements H1–H6
Checked independently from the delivered MP4 with ffprobe.
| H1 | h264 High yuv420p | Pass | |
|---|---|---|---|
| H1 | faststart | Pass | |
| H2 | 1080×1920 SAR1:1 | Pass | 1080x1920 SAR 1:1 |
| H3 | 30 fps | Pass | 30 fps |
| H4 | 55-60 s | Pass | 58.000 s |
| H5 | peak≤ 20Mbps, <200MB | Pass | peak 0.59 Mbps over any 1 s, 2.6 MB |
| H6 | AAC-LC 48k stereo | Pass |
From the agent’s report
Short, verbatim quotes from the REPORT.md the agent wrote itself.
Minimal implementation.
Pedestrians 0 (simplified).
Contact sheet and route map
