Skip to content
Banja Labs
HomeWorkWhat we doThe lab
Let's talk

Traffic light with only the green light lit

One task across 30 published runs. Inspect the scores and outputs together, with each run's execution method shown.

All experimentsModel resultsDesigner buildsCompare models
Task
SVG-0006
Domain
SVG and graphics
Difficulty
medium

Leading published run: claude-opus-4-8 (low reasoning) at 100.0% composite. Lowest: gemini-3.5-flash at 0.0%. 30 published runs shown on this task.

How this task was evaluated

Results are ordered by this task's published score. API runs use one prompt and one completion; agent runs use a coding harness with tools and iteration. Inspect the execution method when comparing results.

Objective scores use the benchmark's execution checks. Judge scores use a vision panel's fixed rubric; the published score leaves out a model's own family. Missing values are labelled rather than treated as zero.

Read the brief

Draw a vertical traffic light in SVG. Show three lamps stacked top to bottom (red, amber, green). Only the green lamp at the bottom is lit (a bright green); the red and amber lamps are dark / off. Use a 40x100 viewBox. Build it from rectangles and circles only - do not use paths, raster images, or text. Write the result to `light.svg`.

Anthropicclaude-opus-4-8
Low reasoning
Agent harness2026-07-21
claude-opus-4-8 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Anthropicclaude-opus-4-8
Medium reasoning
Agent harness2026-07-21
claude-opus-4-8 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Anthropicclaude-opus-4-8
High reasoning
Agent harness2026-07-21
claude-opus-4-8 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Anthropicclaude-opus-4-8
Extra-high reasoning
Agent harness2026-07-21
claude-opus-4-8 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Anthropicclaude-opus-4-8
Max reasoning
Agent harness2026-07-21
claude-opus-4-8 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Anthropicclaude-sonnet-4-6
High reasoning
Agent harness2026-07-21
claude-sonnet-4-6 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Anthropicclaude-sonnet-5
High reasoning
Agent harness2026-07-21
claude-sonnet-5 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Anthropicclaude-fable-5
High reasoning
Agent harness2026-07-09
claude-fable-5 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Anthropicclaude-haiku-4-5
High reasoning
Agent harness2026-07-09
claude-haiku-4-5 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Zhipuglm-5.2
default reasoning
API single-shot2026-07-17
glm-5.2 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Moonshotkimi-k2.7-code
default reasoning
API single-shot2026-07-17
kimi-k2.7-code rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Moonshotkimi-k3
default reasoning
API single-shot2026-07-17
kimi-k3 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
OpenAIgpt-5.5
High reasoning
API single-shot2026-07-17
gpt-5.5 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
OpenAIgpt-5.4-mini
High reasoning
API single-shot2026-06-23
gpt-5.4-mini rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
OpenAIgpt-5.6-sol
High reasoning
API single-shot2026-07-09
gpt-5.6-sol rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
OpenAIgpt-5.6-sol
Max reasoning
API single-shot2026-07-09
gpt-5.6-sol rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
OpenAIgpt-5.6-terra
High reasoning
API single-shot2026-07-09
gpt-5.6-terra rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
OpenAIgpt-5.6-luna
High reasoning
API single-shot2026-07-09
gpt-5.6-luna rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
Googlegemini-3.1-flash-lite
default reasoning
API single-shot2026-06-23
gemini-3.1-flash-lite rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
xAIgrok-4.5
High reasoning
API single-shot2026-07-09
grok-4.5 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
xAIgrok-4.5
Medium reasoning
API single-shot2026-07-09
grok-4.5 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
xAIgrok-4.5
Low reasoning
API single-shot2026-07-09
grok-4.5 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
xAIgrok-4.3
default reasoning
API single-shot2026-07-03
grok-4.3 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
xAIgrok-4.20-reasoning
default reasoning
API single-shot2026-06-25
grok-4.20-reasoning rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
xAIgrok-build-0.1
default reasoning
API single-shot2026-06-25
grok-build-0.1 rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
xAIgrok-composer-2.5-fast
default reasoning
Agent harness2026-06-25
grok-composer-2.5-fast rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
DeepSeekdeepseek-v4-flash
default reasoning
API single-shot2026-06-25
deepseek-v4-flash rendering of the Traffic light with only the green light lit benchmark - composite 100.0%
Open
Composite 100.0%Objective 100.0%
Open outputFull run
DeepSeekdeepseek-v4-pro
default reasoning
API single-shot2026-06-25
deepseek-v4-pro rendering of the Traffic light with only the green light lit benchmark - composite 91.7%
Open
Composite 91.7%Objective 91.7%
Open outputFull run
Googlegemini-3.1-pro-preview
High reasoning
API single-shot2026-06-23
gemini-3.1-pro-preview rendering of the Traffic light with only the green light lit benchmark - composite 50.0%
Open
Composite 50.0%Objective 50.0%
Open outputFull run
Googlegemini-3.5-flash
default reasoning
API single-shot2026-06-23
gemini-3.5-flash rendering of the Traffic light with only the green light lit benchmark - composite 0.0%
Open
Composite 0.0%Objective 0.0%
Open outputFull run

Product thinking,
design and engineering

An independent product lab, building in Australia.

hello@banja.au
HomeWorkWhat we doThe labStart a conversation
About the labProductsExperimentsJournalWebsites & brand? Norex
© 2026 Banja LabsBuilt together in Australia
PrivacyTermsLinkedIn ↗