github.com (opens in a new tab)· Hosted on GitHub

What AI says about AgentBench

AgentBench is a benchmark for evaluating large language models as autonomous agents across diverse environments including operating systems, databases, knowledge graphs, and web tasks.

Since Jul 5Updated Nov 20, 2026Based on 11,055 AI answersReviewed by

How we measure

Mentioned in · Public index

<1%

of AI answers about AgentBench and its rivals · since Jul 5

Based on 11,055 AI answers

  • ChatGPT<1%
  • Google AI Mode<1%

How often AI mentions AgentBench

Share of AI answers about AgentBench and its rivals

Track your brand free to see 13 weeks of history.
Use the left and right arrow keys to read each date.
Sep 21OctNov 9
  • · ChatGPT switched to GPT-6 Luna

Key takeaways for AgentBench

  • Engine split

    ChatGPT (<1%) and Google AI Mode (<1%) mention AgentBench at similar rates.

  • Strongest market

    AgentBench ranks #105 in LLM Observability and Evaluation Platforms (<1%), its highest standing across 2 tracked markets. View LLM Observability and Evaluation Platforms

  • Own-site citations

    arxiv.org is the top external source (27%), while AgentBench's own site accounts for 27% of citations.

Markets where AI mentions AgentBench

All marketsChatGPTGoogle AI Mode

Topics where AI mentions AgentBench

TopicMentioned in · on this topic

AgentBench in AI answers

1 of 3

I am preparing to release a multimodal agent. Who specializes in evaluating "agent trajectory" quality?

ChatGPTChatGPT· Sep 26, 2026
I'd also look at the people/groups behind AgentBench, WebArena, τ-bench, and related agent benchmarks.
Google AI ModeGoogle AI ModeNo answer mentioning AgentBench

AgentBench: common questions

Does AI recommend AgentBench?
AI mentions AgentBench in <1% of tracked answers across ChatGPT and Google AI Mode. ChatGPT mentions AgentBench in <1% of answers and Google AI Mode in <1%.
Who are AgentBench's main competitors in AI answers?
AI most often compares or mentions AgentBench alongside LangSmith, Braintrust, Arize Phoenix and MLflow, including in LLM Observability and Evaluation Platforms.
Which sources does AI cite for AgentBench?
The websites AI cites most often in answers naming AgentBench include arxiv.org, github.com and aitoolskit.io.
For agents: this page as Markdown, parse.gl/brands/github-com-137.md.

This page

Reuse: Figures and tables on this page are free to republish with attribution: a link to https://parse.gl/brands/github-com-137 beside them. A standard followed link (dofollow) is preferred.

Read and search

This brand's shares and rivals, with 95% ranges: GET agent.parse.gl/v1/tools/check_brand?subject=brd_8406cc91-503e-4bd7-a4bf-1d96116d086b
Search brands: GET parse.gl/api/public/search?q=<words>

Connect

Connect an assistant: agent.parse.gl/mcp

An agent can save work before signup: POST https://agent.parse.gl/v1/workspaces

What it costs

No account: Public figures and up to 5 examples per list, about 120 brands a day per address.
Free account: $5 in credit, your brand's first check (20 buyer questions) and a free weekly check.
After your first top-up: Exports and deeper search results. The first top-up starts at $5.
Live checks: $0.01 per question per engine ($0.05 on Gemini and Claude).
Full public answers and pages: 25 free per day, then $0.005 each.
Public answers older than 90 days: $0.001 each the first time your workspace counts or lists one, then yours.