AI Agent Leaderboard. The ranking of rankings, updated weekly.
One benchmark can flatter a model. So we average each model's position across independent agent boards, show where they agree and disagree, and refresh the list every week.
The leaderboard
Lower average rank is better. Select a model to see its score on each board.
As of November 3, 2026, Claude Opus 5.5 holds first place on the overall board, with an average rank of 1.67 across 3 independent boards, ahead of Claude Sonnet 5.5 (2.00).
One board per publisher, so no publisher gets two votes. A model needs a rank on at least 2 of these 3 boards to be listed.
| Rank | Model | Average rank | Arena Agent | Terminal-Bench 4.0 | Vals Index | Agreement | Week |
|---|---|---|---|---|---|---|---|
| 1 |  Anthropic · Proprietary | 1.67 | 1 | 1 | 3 | Boards agree | – |
| 2 | Anthropic · Proprietary | 2.00 | 3 | 2 | 1 | Boards agree | – |
| 3 |  Google · Proprietary | 2.00 | 2 | – | 2 | Boards agree | – |
| 4 | Anthropic · Proprietary | 4.50 | 4 | 5 | – | Boards agree | – |
| 5 |  OpenAI · Proprietary | 4.67 | 5 | 3 | 6 | Mostly agree | – |
| 6 | Google · Proprietary | 5.50 | 7 | – | 4 | Mostly agree | – |
| 7 | OpenAI · Proprietary | 5.67 | 6 | 3 | 8 | Mostly agree | – |
| 8 |  Meta · Proprietary | 7.00 | 9 | – | 5 | Mostly agree | – |
| 9 | Anthropic · Proprietary | 9.00 | 12 | 6 | – | Mostly agree | – |
| 10 | OpenAI · Proprietary | 9.00 | 11 | 7 | – | Mostly agree | – |
| 11 | Anthropic · Proprietary | 9.00 | 10 | 8 | – | Boards agree | – |
| 12 | OpenAI · Proprietary | 9.50 | 8 | – | 11 | Mostly agree | – |
| 13 | Meta · Proprietary | 11.00 | 15 | – | 7 | Boards disagree | – |
| 14 |  Moonshot AI · Proprietary | 11.00 | 13 | – | 9 | Mostly agree | – |
| 15 | OpenAI · Proprietary | 12.50 | 14 | 11 | – | Mostly agree | – |
| 16 |  DeepSeek · Open weights | 14.00 | 18 | – | 10 | Boards disagree | – |
| 17 |  xAI · Proprietary | 15.33 | 21 | 10 | 15 | Boards disagree | – |
| 18 |  Xiaomi · Open weights | 16.50 | 17 | – | 16 | Boards agree | – |
| 19 |  Z.ai · Open weights | 16.67 | 29 | 9 | 12 | Boards disagree | – |
| 20 | Google · Proprietary | 18.00 | 19 | 17 | 18 | Boards agree | – |
| 21 | Anthropic · Proprietary | 18.50 | 16 | 21 | – | Mostly agree | – |
| 22 |  Alibaba · Proprietary | 19.00 | 31 | 13 | 13 | Boards disagree | – |
| 23 | Anthropic · Proprietary | 19.00 | 24 | – | 14 | Boards disagree | – |
| 24 | Meta · Proprietary | 20.00 | 20 | 20 | – | Boards agree | – |
| 25 | OpenAI · Proprietary | 22.00 | 22 | 19 | 25 | Mostly agree | – |
| 26 | xAI · Proprietary | 23.00 | 30 | 16 | – | Boards disagree | – |
| 27 | Xiaomi · Open weights | 23.50 | 28 | – | 19 | Boards disagree | – |
| 28 |  Tencent · Open weights | 24.00 | 26 | – | 22 | Mostly agree | – |
| 29 | Z.ai · Open weights | 24.33 | 35 | 12 | 26 | Boards disagree | – |
| 30 | OpenAI · Proprietary | 24.50 | 34 | 15 | – | Boards disagree | – |
| 31 | Moonshot AI · Open weights | 25.50 | 27 | – | 24 | Mostly agree | – |
| 32 | DeepSeek · Open weights | 27.50 | 32 | – | 23 | Boards disagree | – |
| 33 |  StepFun · Proprietary | 28.00 | 36 | – | 20 | Boards disagree | – |
| 34 | Google · Proprietary | 28.00 | 33 | 23 | – | Boards disagree | – |
| 35 |  MiniMax · Open weights | 29.00 | 37 | – | 21 | Boards disagree | – |
| 36 |  Mistral · Proprietary | 29.50 | 42 | – | 17 | Boards disagree | – |
| 37 |  Reflection · Open weights | 35.00 | 41 | – | 29 | Boards disagree | – |
| 38 |  NVIDIA · Open weights | 40.50 | 51 | – | 30 | Boards disagree | – |
| 39 |  Thinking Machines · Open weights | 42.00 | 53 | – | 31 | Boards disagree | – |
| 40 | Thinking Machines · Open weights | 43.00 | 54 | – | 32 | Boards disagree | – |
The boards behind this ranking
- Arena Agent by ArenaOutcomes from real agent sessions run by people on Arena. 55 models. Updated Oct 28.
- Terminal-Bench 4.0 by Stanford / Harbor / Laude InstituteFrontier-difficulty tasks in a real terminal. 23 models. Updated Sep 29.
- Vals Index by Vals AIAgentic finance, coding, legal and tax work, weighted by US GDP share. 32 models. Updated Oct 30.
20 more models appear on too few of these boards to average fairly, so they are not ranked here.
Sources: Arena leaderboard dataset (CC BY 4.0; we keep each model's best configuration and renumber), Terminal-Bench, Vals AI benchmarks. Re-read every week. How we calculate it. eSEOspace does not run these benchmarks and has not tested these models for this ranking. The order comes from the formula, and no one can pay for a position. eSEOspace is a member of the Claude Partner Network, and Claude is one of the models ranked here.
How to read it
How to read this ranking
Read the average rank first: lower is better. Then look at the Agreement column. A model the boards agree on is a safer bet than one that is first on one board and twentieth on another.
Select a model's name to see its score on each board and the harness used. Use the tabs to switch from the overall list to coding, knowledge work or reliability, and the filters to narrow by lab or to open-weight models.
Other views: Coding. Knowledge work. Reliability. Methodology.
Limits
What this ranking can't tell you
A high average tells you a model did well on these tests. It does not prove it will handle your work, your data or your tools. Run your own trial on real tasks before you commit.
Boards do not cover every model. A model missing from a tough board can look better than one that was tested and scored lower, which is why we show how many boards each model appears on and whether they agree.
Benchmarks can also be gamed. Researchers have shown that agents can sometimes score well by exploiting a test rather than solving it, so treat a small gap between two models as a tie.
FAQ
Common questions
What is the AI Agent Leaderboard?
How often is it updated?
Why average ranks instead of scores?
Is this a ranking of AI models or AI agents?
Can a company pay to be listed or move up?
Where can I get the raw numbers?
Next steps
Go deeper
AI Agent vs Chatbot vs Automation: What Is Different
Read the guide →Model vs Agent vs Harness: Why the Same AI Scores Differently
Read the guide →AI Agent Cost Per Task: Why Token Price Misleads
Read the guide →AI Agent Frameworks Compared: Claude SDK, OpenAI, LangGraph
Read the guide →Best AI Agents for Accounting and Bookkeeping: 9 Options
Read the guide →Best AI Customer Support Agents: 9 Options Compared
Read the guide →All AI agent guides
See all guides →
Project Managers who will work with you on your project!
Irina S.hello@eseospace.com
Benjamin G.ben@eseospace.com
David G.david@eseospace.comWork with us
Talk to the team behind your project
Tell us what you need and we’ll match you with the right people.
Book a Discovery Call
Jeanette M.jeanette@eseospace.com
Mark S.mark@eseospace.com
Maryna P.maryna@eseospace.com
Wondering what this costs? Every service has published pricing, and you don't need a discovery call to see it.
View pricingPicking an AI agent for your business?
A benchmark rank tells you which models are strong. It does not tell you which one fits your data, your tools and your risk. Tell us what you want the agent to do and we will give you a straight answer.
Talk to our team
