Lava lamp v1 · Claude Code · run 7250efd0 · 07.10.2026 21:39 Tallinn

Claude Haiku 5.5

Max effort

Works

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

Compare →Transcript Source

Desktop capture of Claude Code (Haiku 5.5)'s output on Lava lamp v1, max effort, 07.10.2026 21:39 (run 7250efd0) — **Desktop** · 612.7 KB · 32cf29037908…

Phone capture of Claude Code (Haiku 5.5)'s output on Lava lamp v1, max effort, 07.10.2026 21:39 (run 7250efd0) — **Phone** · 207.5 KB · 9d74b6cc980e…

Motion capture of Claude Code (Haiku 5.5)'s output on Lava lamp v1, max effort, 07.10.2026 21:39 (run 7250efd0) — **Motion** · 615.2 KB · d0fdd75dde43…

Build time
20m 00s
8th fastest of 12 models at max effort. Wall time, start to graded: 21m 00s.
Output tokens
263,901
10th fewest of 12 models at max effort.
Output speed
220 tok/s
Highest of 12 models at max effort.
API cost
$1.37
The CLI’s list-price estimate, not a charge.

Along the row

Claude Haiku 5.5 at every effort.

Down the column

Other models at max effort.

Fastest median first · this run: 20m 00s

Hidden checks

7 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 3c2937dcf74727c405d68f6a4cd81cac053116c0caa72c2c59733a0d0fceaf78, 1074.9 KB.

Run ID: 7250efd0-831e-43a6-9e0b-380edc8bf4c9
Benchmark: lava-lamp v1
Experiment: Single run
Started: 07.10.2026 21:39 Tallinn
Graded: revision 1 · 07.10.2026 22:00

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	fail
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-haiku-5-5	claude-haiku-5-5	harness result, 58 of 58 API requests
Effort	max	max	57 of 58 API requests
Harness	not recorded	2.1.293 (Claude Code)	Version as the CLI reported it; binary SHA-256 4e21122a2278…
Deadline	20 min	21m 00s used	Stopped at the time limit

Tool decisions

57 calls · 56 accepted · 1 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.

Bash25 calls · 24 accepted · 1 rejected · 24 failed
Edit18 calls · 18 accepted
Read7 calls · 7 accepted
Write5 calls · 5 accepted
Grep2 calls · 2 accepted

Telemetry

Time and usage.

Generation time: 20m 00s
First content: 1.0 s
Output speed: 220.0 tokens/s, whole run including tools
API-equivalent cost: $1.37 (list-price estimate, not a charge)
Actual charge: not reported
Billing: subscription

Input tokens

13,204,160 tokens

Cache reads12,909,67898%
Cache writes294,3682.2%
Uncached input114<0.1%

Output tokens

263,901 tokens

Reasoning226,78686%
Other output37,11514%

Usage is incomplete; missing categories stay unknown. 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: 137.3 KB, 57 redactions. Download transcript.txt

Source files

1 file · 19.6 KB

Files in the output bundle
File	Size	SHA-256	Redactions
index.html	19.6 KB	18c6eb05a1d7f6a4a3cf3e1fb7d9eff07223169a7d0128cdd172319f2c054b79	0

Bundle SHA-256 18c6eb05a1d7f6a4a3cf3e1fb7d9eff07223169a7d0128cdd172319f2c054b79

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)	07.10.2026 22:00	works

Redaction: automated scan run; checked by Jaanus on 22.10.2026 15:37. Redacted: 25 local-path, 32 username.

Download run.json