CmdBenchby CmdBrief
Menu

Lava lamp v1 · Codex CLI · run 5a62dd4e · 28.10.2026 08:44 Tallinn

GPT-5.6 Sol

Max effort

Does not work

6 of 8 checks passed. 2 of 5 blocker checks failed.

Compare →TranscriptSource
Desktop capture of Codex CLI (GPT-5.6 Sol)'s output on Lava lamp v1, max effort, 28.10.2026 08:44 (run 5a62dd4e)
Desktop · 14.8 KB · f0074cac4917…
Phone capture of Codex CLI (GPT-5.6 Sol)'s output on Lava lamp v1, max effort, 28.10.2026 08:44 (run 5a62dd4e)
Phone · 6.2 KB · f2bb02b35dbd…
Motion capture of Codex CLI (GPT-5.6 Sol)'s output on Lava lamp v1, max effort, 28.10.2026 08:44 (run 5a62dd4e)
Motion · 14.8 KB · f0074cac4917…

Along the row

GPT-5.6 Sol at every effort.

  1. GPT-5.6 Sol at low effort: desktop capture of run 184f5448 Low1m 46s

    Works

    1 run
  2. GPT-5.6 Sol at medium effort: desktop capture of run d3591876 Medium2m 20s

    Fails

    1 run
  3. GPT-5.6 Sol at high effort: desktop capture of run 456bc2ff High2m 34s

    Works

    1 run
  4. GPT-5.6 Sol at extra high effort: desktop capture of run 0e1f62c1 Extra high4m 05s

    Fails

    1 run
  5. GPT-5.6 Sol at max effort: desktop capture of run 5a62dd4e Max7m 53s

    Fails

    This effort

Down the column

Other models at max effort.

Fastest median first · this run: 7m 53s

  1. GPT-5.6 Luna at max effort: desktop capture of run ab2b3d6aGPT-5.6 Luna2m 22s

    Works

    3.3× faster than this run
  2. GPT-5.6 Terra at max effort: desktop capture of run f8748487GPT-5.6 Terra4m 04s

    Fails

    1.9× faster than this run
  3. GPT-6 Luna at max effort: desktop capture of run 8dee2af3GPT-6 Luna4m 38s

    Works

    1.7× faster than this run
  4. GPT-6.1 Terra at max effort: desktop capture of run f1cd6f87GPT-6.1 Terra4m 55s

    Works

    1.6× faster than this run
  5. GPT-6 Sol at max effort: desktop capture of run f796ca14GPT-6 Sol13m 48s

    Works

    1.8× slower than this run
  6. GPT-6.1 Sol at max effort: desktop capture of run aa82c8e5GPT-6.1 Sol16m 56s

    Works

    2.1× slower than this run
  7. Claude Haiku 5.5 at max effort: desktop capture of run 7250efd0Claude Haiku 5.520m 00s

    Works

    2.5× slower than this run
  8. GPT-6 Astra at max effort: desktop capture of run 9b8c977cGPT-6 Astra36m 32s

    Works

    4.6× slower than this run
  9. Claude Fable 5.1 at max effort: desktop capture of run 1d1e1f7fClaude Fable 5.136m 52s

    Works

    4.7× slower than this run
  10. Claude Opus 5.5 at max effort: desktop capture of run e684fb53Claude Opus 5.566m 39s

    Works

    8.5× slower than this run
  11. Claude Sonnet 5.5 at max effort: desktop capture of run 42c0200cClaude Sonnet 5.587m 03s

    Works

    11.0× slower than this run

Hidden checks

6 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 detailsagent config, what ran, tokens, source and revisions
  • Video (video): not published, videos are withheld. SHA-256 8f3d2b2379f20873cc34cc047d9de43fd42a48f69dca6340135832b8659aa711, 105.7 KB.
Run ID
5a62dd4e-ede7-4aaf-ad13-2713ac29c230
Benchmark
lava-lamp v1
Experiment
Single run
Started
28.10.2026 08:44 Tallinn
Graded
revision 1 · 28.10.2026 08:53

Works and checks, in full

2 of 5 blocker checks failed.

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.

Contract v1 checks for this run, with blocker and result
RequirementCheckBlockerResult
none (v1)Self-contained index.html is presentBlockerpass
none (v1)Visible page contentBlockerpass
none (v1)Visible autonomous animationBlockerfail
none (v1)Accessible on/off controlExtrapass
none (v1)390 px mobile layout has no horizontal overflowExtrapass
none (v1)Local animation frame performanceExtrapass
none (v1)No browser errorsBlockerfail
none (v1)No attempted external or linked-resource requestsBlockerpass

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.

Requested agent config beside what the run reported
FieldRequestedDetectedEvidence
Modelgpt-5.6-solnot reportedThe harness reported no model.
Effortmaxnot reportedThe harness reported no effort; the requested effort is shown.
Harnessnot recordedcodex-cli 0.165.0Version as the CLI reported it; binary SHA-256 a45099154d7f…
Deadline60 min8m 52s usedCompleted

Tool decisions

10 calls · 10 without a decision

The harness reports no permission decision for its tool calls; the result of each call is still recorded.

  • command_execution9 calls
  • file_change1 call

Telemetry

Time and usage.

Generation time
7m 53s
First content
5.4 s
Output speed
58.3 tokens/s, whole run including tools
API-equivalent cost
not reported
Actual charge
not reported
Billing
subscription

Input tokens

348,182 tokens
  • Cache reads336,64097%
  • Cache writesnot reported
  • Uncached input11,5423.3%

Output tokens

27,571 tokens
  • Output27,571100%
  • 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: 6.6 KB, 1 redaction. Download transcript.txt

Source files

1 file · 35.2 KB
Files in the output bundle
FileSizeSHA-256Redactions
index.html35.2 KB2c2313e98f196678f0f50052381c449cca0c2f3ed58e52eec32a80be504877090
Bundle SHA-256 2c2313e98f196678f0f50052381c449cca0c2f3ed58e52eec32a80be50487709

Revision history

This page shows revision 1.

Grades are append-only: a regrade adds a revision and never rewrites an earlier one.

Grading revisions of this run
RevisionGradedWorks
1 (current)28.10.2026 08:53does not work

Redaction: automated scan run; checked by Jaanus on 28.10.2026 17:51. Redacted: 1 username.

Download run.json