Unprompted

Operated by Skald Studio, which sells AI visibility work. No placement on this chart is for sale.

AI Coding Assistants · 2026-09-14

AI names Claude Code first in 31% of runs.

Every week we ask the AI assistants the same 15 buying questions, 5 times each, and publish exactly who they name. The questions, the method and every raw answer are public.

See the full board →

01ask(“What is the best AI coding assistant?”)
02ask(“Which AI is best for writing code in 2026?”)
03ask(“What AI coding tool should a beginner start with?”)
04ask(“Which AI coding assistant is best for a large existing codebase?”)
05ask(“What is the best AI pair programmer for a professional developer?”)
06ask(“Which AI coding tool gives the best value for money?”)
07ask(“What is the best AI agent for autonomous coding tasks?”)
08ask(“Which AI coding assistant works best in the terminal?”)
09ask(“What is the best AI code editor?”)
10…6 more
RUN 2026-09-14METHOD v25 RUNS/Q5 ENGINES: CHATGPT · CLAUDE · CLAUDE-CODE · CODEX · PERPLEXITY
MEASURED 2026-09-14UPDATES EVERY MONDAYCHECK THE RAW DATA →

The methodology version changed; movement is not comparable.

15 questions · hover a step to read one down the board
Brand
01Claude Code249/37431%
02GitHub Copilot291/37430%
03Cursor314/37424%
05CodeRabbit25/3742%
06Devin33/3742%
08Replit Agent31/3741%
09Greptile18/3741%
10Codex157/3741%
11Qodo14/3741%
12Windsurf76/3740%
13Amazon Q61/3740%
14JetBrains AI58/3740%
15Aider71/3740%
16Cline52/3740%
17Tabnine35/3740%
18Zed24/3740%
19Continue23/3740%
20Augment22/3740%
21OpenCode13/3740%
22Roo Code12/3740%
23OpenHands11/3740%
24Playcode AI2/3740%

Why we repeat each question

These systems do not give the same answer twice. Ask the same question on Monday and Wednesday and you can get different brands. Asking once and publishing the result would be publishing a coin flip, so we ask repeatedly and report how often a brand was named. Each step on the board above is one question, and its height is how often that brand was named for it.

Read the full method →