Lava lamp v1 · Claude Code · run 483f29b0 · 30.09.2026 19:41 Tallinn

Claude Opus 5.5

Medium effort

Works

8 of 8 checks passed. All 5 blocker checks passed.

Compare →Transcript Source

Desktop capture of Claude Code (Opus 5.5)'s output on Lava lamp v1, medium effort, 30.09.2026 19:41 (run 483f29b0) — **Desktop** · 629.3 KB · c8fc40bc194e…

Phone capture of Claude Code (Opus 5.5)'s output on Lava lamp v1, medium effort, 30.09.2026 19:41 (run 483f29b0) — **Phone** · 142.9 KB · f8ca9963d52e…

Motion capture of Claude Code (Opus 5.5)'s output on Lava lamp v1, medium effort, 30.09.2026 19:41 (run 483f29b0) — **Motion** · 629.8 KB · 653ead796163…

Build time
8m 56s
11th fastest of 11 models at medium effort. Wall time, start to graded: 9m 58s.
Output tokens
54,514
10th fewest of 11 models at medium effort.
Output speed
102 tok/s
6th highest of 11 models at medium effort.
API cost
$2.01
The CLI’s list-price estimate, not a charge.

Along the row

Claude Opus 5.5 at every effort.

Down the column

Other models at medium effort.

Fastest median first · this run: 8m 56s

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 detailsagent config, what ran, tokens, source and revisions

Video (video): not published, videos are withheld. SHA-256 ed7e7a375dba72ea2d59e163e8b69f31dd4b94c61d4080353b8a1d101e54c938, 1426.0 KB.

Run ID: 483f29b0-33e4-465e-a4ac-66e6a6339f86
Benchmark: lava-lamp v1
Experiment: Effort pair
Started: 30.09.2026 19:41 Tallinn
Graded: revision 1 · 30.09.2026 19:51

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.

Contract v1 checks for this run, with blocker and result
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.

Requested agent config beside what the run reported
Field	Requested	Detected	Evidence
Model	claude-opus-5-5	claude-opus-5-5	harness result, 20 of 20 API requests
Effort	medium	medium	20 of 20 API requests
Harness	Claude CLI	2.1.285 (Claude Code)	Version as the CLI reported it; binary SHA-256 51f09bd1e021…
Deadline	10 min	9m 58s used	Completed

Tool decisions

38 calls · 38 accepted · 0 rejected

Each tool call is allowed or blocked by the harness’s permission settings before it runs. Allowed is not the same as succeeded: an allowed call can still fail.

Bash18 calls · 18 accepted · 14 failed
Edit16 calls · 16 accepted
Write4 calls · 4 accepted

Telemetry

Time and usage.

Generation time: 8m 56s
First content: 2.0 s
Output speed: 101.7 tokens/s, whole run including tools
API-equivalent cost: $2.01 (list-price estimate, not a charge)
Actual charge: not reported
Billing: subscription

Input tokens

1,227,560 tokens

Cache reads1,141,14193%
Cache writes86,3797.0%
Uncached input40<0.1%

Output tokens

54,514 tokens

Reasoning31,97859%
Other output22,53641%

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: 95.5 KB, 25 redactions. Download transcript.txt

Source files

1 file · 20.7 KB

Files in the output bundle
File	Size	SHA-256	Redactions
index.html	20.7 KB	83e6eb6e1b82279184ab79aa9a73e771842c070da8fb486efa0fa07a6cf27d20	0

Bundle SHA-256 83e6eb6e1b82279184ab79aa9a73e771842c070da8fb486efa0fa07a6cf27d20

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
Revision	Graded	Works
1 (current)	30.09.2026 19:51	works

Redaction: automated scan run; checked by Jaanus on 22.10.2026 15:37. Redacted: 15 local-path, 3 user-path, 7 username.

Download run.json