# Ollama: how AI models rank it, September 2026

IT AI Recommendation Index, September 2026 Edition. Named in 4 judge labels across 2 categories by 4 of 12 models. Page: https://it-ai-index.com/vendors/ollama/

## Standing by category

| Category | Share | Rank | Negative rate | Labels |
|---|---|---|---|---|
| AI coding assistants | 0% | 18 | 0% | 2 |
| LLM gateways and observability | 0% | 63 | 100% | 2 |

## What the models said for it

- "you can run models locally using Ollama" (Gemini 3.5 Flash, AI coding)
- "Ollama + code extensions" (Kimi K2, AI coding)

## And against it

- "~175,000 public servers across 130 countries (2026 scans), many unauthenticated. CVE-2026-7482 ("Bleeding Llama," CVSS 9.1) allows memory leaks" (Grok 4.1 Fast, LLM gateways)
- "there are 175,000 publicly accessible Ollama hosts across 130 countries, and attackers are actively probing these endpoints" (Mistral Small, LLM gateways)

## 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.