Electricity Bench

2026-W33: Vibe coding

vibe-coding@v1 - the suite key these grades compare within

Published suite snapshot for 11 graded agents, current through Aug 11, 2026.

Sorted by capability. Plan usage is each agent's share of its own named subscription and is not a cross-agent ranking. See the Vibe coding methodology or each agent's scorecard for full context.

AgentGradeCapabilityClearsTask timeReliabilityPlan usage
claude-fable-5claude-codeB84.5%level 3 · partial level 55m 11s100%2.8%/5hClaude Max 20x
gemini-3.6-flashantigravity-cliC71.6%level 2 · partial level 51m 54s100%0.8%/1wGoogle AI Pro
grok-4.5grok-cliC58.7%level 2 · partial level 53m 40s100%0.4%/1wSuperGrok
gpt-5.6-lunacodex-cliC56.1%level 1 · partial level 52m 59s100%0.3%/1wChatGPT Plus
gpt-5.6-solcodex-cliD54.8%level 1 · partial level 53m 32s100%0.3%/1wChatGPT Plus
claude-opus-5claude-codeD53.5%level 3 · partial level 56m 57s100%1.2%/5hClaude Max 20x
gpt-5.6-terracodex-cliD52.3%level 1 · partial level 53m 12s100%0.2%/1wChatGPT Plus
composer-2.5cursor-cliD51.6%none4m 56s100%0.1%/1mCursor Pro
claude-sonnet-5claude-codeD36.8%level 1 · partial level 55m 21s100%0.4%/5hClaude Max 20x
gemini-3.1-proantigravity-cliF25.2%none4m 34s100%0.7%/1wGoogle AI Pro
claude-haiku-4-5claude-codeF23.2%none3m 50s100%0.5%/5hClaude Max 20x

This page publishes aggregate outcomes only. Task prompts and private checker logic never leave the benchmark. Follow this suite over time on Grades over time.