# Confluence vs BookStack: which do AI models recommend for knowledge base, September 2026

GTM AI Recommendation Index, September 2026 Edition, Knowledge base and self-service. One of twelve models named Confluence first on the direct prompt; zero named BookStack. Page: https://gtm-ai-index.com/customer/knowledge-base-and-self-service/confluence-vs-bookstack/

| | First-choice share | Rank | Negative rate | Labels | Models naming it |
|---|---|---|---|---|---|
| Confluence | 6% | #4 of 16 | 14% | 44 | 12 of 12 |
| BookStack | 6% | #6 of 16 | 15% | 13 | 10 of 12 |

## The direct prompt, model by model

- Kimi K2: confluence first (first choices: Confluence) (alternatives: Guru, Helpjuice, Notion, Slite, Zendesk Guide)
- GPT-5.4 mini: neither first, one named (first choices: Zendesk) (alternatives: Confluence, Intercom)
- Gemini 3.5 Flash: neither first, one named (first choices: Document360, Guru) (alternatives: Confluence, Help Scout Docs, Helpjuice, Notion)
- Perplexity Sonar: neither first, one named (first choices: Document360) (alternatives: Confluence, Helpjuice, Zendesk Guide)
- Grok 4.1 Fast: neither first, one named (first choices: Document360) (alternatives: Confluence, Helpjuice, Notion, Slite, Zendesk Guide)
- Mistral Small: neither first, one named (first choices: Plain) (alternatives: Confluence, Document360, Freshdesk, Guru, Slite)
- GLM 4.7 FlashX: neither first, one named (first choices: Helply) (alternatives: Confluence, Notion, Plain, Slite)
- Claude Haiku 4.5: neither named (alternatives: Bloomfire, Document360, Lore, Stonly, Zendesk Guide)
- DeepSeek V4 Flash: neither named (first choices: Document360) (alternatives: Freshdesk, Guru, Help Scout, Helpjuice, Zendesk)
- Llama 4 Maverick: neither named (first choices: Helpjuice) (alternatives: Document360, Freshdesk, Pylon)
- Qwen 3.7 Flash: neither named (first choices: Guru, Intercom) (alternatives: Atlassian Confluence, Document360, Zendesk)
- MiniMax M2.5: neither named (first choices: Document360) (alternatives: Helpjuice, Zendesk Guide/Knowledge)

## What the models said about Confluence

- "SharePoint and Confluence excel at document lifecycle management and collaboration, but their knowledge access features are basic and not purpose-built for customer experience delivery" (Claude Haiku 4.5, comparative prompt, soft negative)
- "legacy wikis like Confluence without updates ... often contain stale, duplicate, or conflicting data" (Grok 4.1 Fast, negative prompt, soft negative)
- "Confluence is incredibly robust but can feel slow, bloated, and clunky to non-technical users." (Gemini 3.5 Flash, negative prompt, soft negative)
- "The top overall choice of knowledge base software for IT teams with tight budgets is Confluence" (Claude Haiku 4.5, budget prompt, first choice)
- "Notion (SMB flexibility) or Confluence (engineering/Atlassian) are the safe defaults" (DeepSeek V4 Flash, comparative prompt, first choice)
- "toward a dedicated platform like Notion, Confluence, or Document360" (Qwen 3.7 Flash, negative prompt, first choice)

## What the models said about BookStack

- "High-Risk Tools to Avoid ... BookStack ... Performance degrades significantly with large interconnected databases" (Mistral Small, negative prompt, hard negative)
- "these are solid but require technical resources to manage properly" (MiniMax M2.5, negative prompt, soft negative)
- "Choose BookStack if you have access to an IT administrator, want to host your own data, and want a completely free tool with no user limits." (Gemini 3.5 Flash, budget prompt, first choice)
- "BookStack offers the best balance of being completely free, easy to use, and requiring minimal technical overhead" (Mistral Small, budget prompt, first choice)
- "If you have technical resources: Start with BookStack or Outline (self-hosted) - lowest long-term cost" (GLM 4.7 FlashX, budget prompt, first choice)

Share is the count of first choices across the direct, paraphrase, budget and scale prompts over all twelve models, for a mid-market B2B company; rank is within the category. Comparisons are drawn for the top eight products in each category. Published under CC BY 4.0; the output is the models' output, and nothing here is a recommendation by the index.