# Claude Code: how AI models rank it, September 2026

IT AI Recommendation Index, September 2026 Edition. Named in 28 judge labels across 1 categories by 12 of 12 models. Page: https://it-ai-index.com/vendors/claude-code/

## Standing by category

| Category | Share | Rank | Negative rate | Labels |
|---|---|---|---|---|
| AI coding assistants | 0% | 13 | 14% | 28 |

## What the models said for it

- "Claude Code is the top-ranked AI coding tool in 2026, powered by Opus 4.6, scoring 80.8% on SWE-bench Verified" (Claude Haiku 4.5, AI coding)
- "*top-performing terminal agent* ... Currently tops SWE-bench (~80%)" (DeepSeek V4 Flash, AI coding)
- "Tools like Cursor (~$20/user) or Claude Code score higher on context awareness and autonomy" (MiniMax M2.5, AI coding)
- "add Claude Code as a metered second tool for the minority doing agentic work" (Llama 4 Maverick, AI coding)

## And against it

- "Also affected by IDEsaster; 100% of tested tools vulnerable to attack chains" (GLM 4.7 FlashX, AI coding)
- "Approach with Extreme Caution: "Agentic" Coding Assistants ... (such as Claude Code, Devin, Cursor, and Windsurf)" (Gemini 3.5 Flash, AI coding)
- "Over-eagerness: ... "fulfill requests too eagerly" — making changes you didn't ask for" (DeepSeek V4 Flash, AI coding)
- "General Risks Across Most (e.g., Cursor, Claude Code, Replit AI)" (Grok 4.1 Fast, AI coding)

## Record

- Method: https://it-ai-index.com/methodology/
- Raw judge labels and full responses: https://it-ai-index.com/data/
- License: CC BY 4.0. Cite as IT AI Recommendation Index, September 2026 Edition, it-ai-index.com.