Subcategory · AI Citation Index
Agent Frameworks
Agent frameworks is a split-verdict category. LangChain surfaces across every engine in 82% of discovery prompts; CrewAI and AutoGen each appear in 75% and 74% respectively, also on all four engines. But when buyers ask AI to compare options head-to-head, CrewAI and AutoGen each win the majority of matchups, while LangChain wins fewer than it loses. LlamaIndex rounds out the consensus tier at 66% discovery share across all engines and a mid-tier win rate. This is a consolidated shortlist with an unusual tension: the brand that captures the most AI attention in discovery doesn't win the most head-to-heads in evaluation.
164 discovery queries · 168 head-to-heads · refreshed Aug 16, 2026
Discovery stage
The shortlist
Across 164 buyer-style "Agent Frameworks" queries
LangChain shows up in 82% of buyer queries about agent frameworks and surfaces across ChatGPT, Claude, Gemini, and Perplexity. CrewAI and AutoGen follow at 75% and 74% respectively, both visible on all four engines. LlamaIndex appears in 66% of queries, also across every engine we track. The next tier drops sharply: handoff.ai and base.ai each surface across all four engines but in fewer than 30% of queries.
Hover or click a logo to see brand details
Get weekly AI visibility changes for Agent Frameworks sent to your inbox.
Score shifts, new entrants, citation gaps — every Monday.
Signal by intent
By topic
Top 5 most-cited brands per intent cluster. Brands with zero citations in a topic are not shown.
Evaluation stage
Head-to-head
How often AI cites each brand across uniform category evaluation prompts · median 3/100
When buyers ask AI to compare agent frameworks, CrewAI and AutoGen tie for the most head-to-head wins, each across 35 comparison queries. LangChain appears in the same number of matchups but wins fewer than it loses. LlamaIndex holds the fourth slot with a mid-tier win rate across 31 comparisons. Microsoft Copilot Studio and UiPath each show up in head-to-heads but lose the majority of those matchups.
Hover or click a logo to see brand details
Each brand's score is the share of category evaluation prompts where AI cited them across all four engines — the same prompt pool for every brand. Brands above the median citation rate have stronger presence in evaluation-stage queries.
Brands to know
In this category
LangChain
Consensus pickPython library for chaining LLM calls
Read brand profile →CrewAI
Consensus pickRole-based multi-agent orchestration framework
Read brand profile →AutoGen
Consensus pickMicrosoft framework for multi-agent conversations
Read brand profile →LlamaIndex
Data framework for LLM applications
Read brand profile →Citation sources
Where AI pulls citations from
1000 citations captured across Agent Frameworks prompt runs.
Vendor pages
286Product, help, and marketing pages from tracked vendors
Independent sources
309Reviews, encyclopedias, forums, press — not vendor-owned
Buyer questions
What AI cites for top Agent Frameworks questions
Buyers ask AI for agent framework recommendations filtered by role, use case, and architecture constraints — phrasings like 'what do system architects prefer for designing robust agent frameworks', 'which agent frameworks are best for data-driven decision making', 'which agent framework tools enhance user interaction'. A smaller set digs into interoperability and customization trade-offs, asking 'what are the interoperability concerns with agent frameworks', 'what role does customization play in agent frameworks selection'.
Discovery
Buyers exploring the categoryEvaluation
Buyers comparing optionsWant to know if AI cites your brand for Agent Frameworks?
Free audit. ChatGPT, Perplexity, Gemini, Claude.
Run an audit →