Lava lamp v1 · Codex CLI · run 61c1a78e · 28.10.2026 10:18 Tallinn
GPT-6 Luna
Extra high effortWorks
8 of 8 checks passed. All 5 blocker checks passed.
What it built: 3 captures
Build time
4m 24s
5th fastest of 12 models at extra high effort. Wall time, start to graded: 5m 23s.
Output tokens
26,339
6th fewest of 12 models at extra high effort.
Output speed
100 tok/s
5th highest of 12 models at extra high effort.
API cost
Not reported
Codex CLI does not report a list-price estimate.
Along the row
GPT-6 Luna at every effort.
Down the column
Other models at extra high effort.
Fastest median first · this run: 4m 24s
GPT-6.1 Terra1m 15sWorks
3.5× faster than this run
GPT-5.6 Luna1m 25sWorks
3.1× faster than this run
GPT-5.6 Terra3m 50sWorks
1.1× faster than this run
GPT-5.6 Sol4m 05sFails
1.1× faster than this run
GPT-6 Sol7m 11sWorks
1.6× slower than this run
Claude Haiku 5.58m 28sWorks
1.9× slower than this run
GPT-6.1 Sol11m 37sWorks
2.6× slower than this run
Claude Sonnet 5.521m 18sWorks
4.8× slower than this run
GPT-6 Astra25m 55sWorks
5.9× slower than this run
Claude Opus 5.528m 09sWorks
6.4× slower than this run
Claude Fable 5.132m 58sWorks
7.5× slower than this run
Hidden checks
8 of 8 checks passed.
- Self-contained index.html is presentblocker
- Visible page contentblocker
- Visible autonomous animationblocker
- Accessible on/off controlextra
- 390 px mobile layout has no horizontal overflowextra
- Local animation frame performanceextra
- No browser errorsblocker
- No attempted external or linked-resource requestsblocker
It works when every blocker passes. How it looks is reviewed blind: pending.
Run details
- Video (video): not published, videos are withheld. SHA-256 0d0b6df40cc5fcf543ff04bf2d9d3210fad42a16f5eecddc00ac6a523879b66c, 812.9 KB.
- Run ID
- 61c1a78e-b931-4b24-b932-0a6f6d91569c
- Benchmark
- lava-lamp v1
- Experiment
- Single run
- Started
- 28.10.2026 10:18 Tallinn
- Graded
- revision 1 · 28.10.2026 10:23
Works and checks, in full
All 5 blocker checks passed.
The rule: a run works when every blocker check passes. If any blocker fails it does not work. If none fails but one has no result yet, it stays pending, never counted as a pass or a fail. Extra checks are reported but do not decide it. Works is not a judgement of looks: that is the visual review.
This is a legacy contract v1 run: the task has no public requirements, and these are the original lava-lamp checks, not a versioned hidden verifier. It is never compared with contract v2 runs.
| Requirement | Check | Blocker | Result |
|---|---|---|---|
| none (v1) | Self-contained index.html is present | Blocker | pass |
| none (v1) | Visible page content | Blocker | pass |
| none (v1) | Visible autonomous animation | Blocker | pass |
| none (v1) | Accessible on/off control | Extra | pass |
| none (v1) | 390 px mobile layout has no horizontal overflow | Extra | pass |
| none (v1) | Local animation frame performance | Extra | pass |
| none (v1) | No browser errors | Blocker | pass |
| none (v1) | No attempted external or linked-resource requests | Blocker | pass |
Verification completed; eligible for benchmark results.
Visual review
Pending.
Pending: the captures have not been reviewed blind yet. Checks decide whether it works; how it looks is decided only in that review.
What ran
Requested and detected.
Detected values come from the run’s own output, never from our settings.
| Field | Requested | Detected | Evidence |
|---|---|---|---|
| Model | gpt-6-luna | not reported | The harness reported no model. |
| Effort | xhigh | not reported | The harness reported no effort; the requested effort is shown. |
| Harness | not recorded | codex-cli 0.165.0 | Version as the CLI reported it; binary SHA-256 a45099154d7f… |
| Deadline | 60 min | 5m 23s used | Completed |
Tool decisions
The harness reports no permission decision for its tool calls; the result of each call is still recorded.
- command_execution5 calls
- file_change4 calls
Telemetry
Time and usage.
- Generation time
- 4m 24s
- First content
- 4.8 s
- Output speed
- 99.6 tokens/s, whole run including tools
- API-equivalent cost
- not reported
- Actual charge
- not reported
- Billing
- subscription
Input tokens
- Cache reads281,85678%
- Cache writesnot reported
- Uncached input80,72922%
Output tokens
- Output26,339100%
- Reasoningnot reported
Usage as the agent reported it at the end of the run. Cache reads are part of input; reasoning is part of output.
Transcript and source
Plain-text downloads.
The redacted transcript and the frozen output bundle, with LF line endings. Hashes are of the original files.
Redacted agent transcript: 15.8 KB, 0 redactions. Download transcript.txt
Source files
| File | Size | SHA-256 | Redactions |
|---|---|---|---|
| index.html | 23.5 KB | 16f2fa359d551a46a101c0e51dab31745b4bcc82bdb61ed8c798569300b95363 | 0 |
Revision history
This page shows revision 1.
Grades are append-only: a regrade adds a revision and never rewrites an earlier one.
| Revision | Graded | Works |
|---|---|---|
| 1 (current) | 28.10.2026 10:23 | works |
Redaction: automated scan run; checked by Jaanus on 28.10.2026 17:51. Nothing needed redacting.