Electricity Bench

2026-W33: Vibe coding

Week 33 of 2026 graded 11 coding agents on vibe coding. Claude Fable 5.1 leads with B. Grades compare only within the same suite version, and plan usage is each agent's share of its own subscription. Every agent ran the same suite cut headless through its own tools.Measured

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

Published suite snapshot for 11 graded agents, current through Aug 14, 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 5.1claude-codeB84.5%level 3 · partial level 55m 11s100%11.2%/5hClaude Max 5x
Grok 4.6grok-cliC+74.2%level 3 · partial level 57m 33s100%0.4%/1wSuperGrok Plus
Gemini 3.8 Flashantigravity-cliC+71.6%level 2 · partial level 53m 37s100%0.7%/1wGoogle AI Pro
GPT-6 Lunacodex-cliD+51.0%level 1 · partial level 52m 59s100%0.1%/1wChatGPT Plus
GPT-6 Solcodex-cliD44.5%level 1 · partial level 53m 32s100%0.8%/1wChatGPT Plus
Claude Opus 5.5claude-codeD43.2%level 3 · partial level 56m 57s100%4.8%/5hClaude Max 5x
GPT-5.6 Terracodex-cliD41.9%level 1 · partial level 53m 12s100%0.1%/1wChatGPT Pro 5x
Composer 2.5cursor-cliD-41.3%none4m 56s100%0.1%/1mCursor Pro
Claude Sonnet 5claude-codeF34.2%level 1 · partial level 55m 21s100%1.6%/5hClaude Max 5x
Claude Haiku 4.5claude-codeF23.2%none3m 50s100%2%/5hClaude Max 5x
Gemini 3.1 Proantigravity-cliF21.3%none4m 34s100%0.7%/1wGoogle AI Pro

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.