Skip to content
New: free AI crawler checker. See if ChatGPT and Claude can read your site. Check now
Get a free audit
Get a free audit Contact us +91 96211 05242

Popular

ResearchUpdated 23 Oct 2026

77.3%success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026)

77.3% success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026) (2026, Stanford HAI, AI Index Report)

ResearchChecked at source, 23 Oct 2026All sizesGlobalAI agents · Capability
Embed CSV JSON Source

AI agent success on benchmark tasks (Stanford AI Index 2026)

GroupEarlierLatest
OSWorld (computer tasks)12%66%
Terminal-Bench (real-world tasks)20%77.3%
Source: Stanford HAI, AI Index Report Benchmark scores reported by the AI Index 2026; each row is one benchmark, an earlier score and the latest one (Terminal-Bench: 2025 to 2026; OSWorld: start year not given). Lab tests, not adoption.digitaltreed.com/statistics

In plain words

What this number really says

77.3% success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026) (2026).

77.3%

the latest figure (2026)

Research

type of source

Why you can trust it. Stanford HAI, AI Index Report: research, research firm or analyst compilation.

Key facts

  • Period 2026
  • Region Global
  • Company size All sizes
  • Next data Yearly

Source: Stanford HAI, AI Index Report, 2026

Method and use

How it was measured, and how to use it

How it was measured
Publisher
Stanford HAI, AI Index Report
Source type
Research: Research firm or analyst compilation
What it measures
77.3% success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026)
Data period
2026 (ends 31 Mar 2026)
Published by the source
13 Apr 2026
In the source's words
the success rate of agents handling real-world tasks improved from 20% in 2025 to 77.3% today, according to Terminal-Bench
Checked at source
23 Oct 2026: Read on hai.stanford.edu (news: "Inside the AI Index: 12 Takeaways from the 2026 Report", 13 Apr 2026) on 2026-10-23. Benchmark score.
Reuse
CC BY-ND 4.0 for the report; check before reuse. Compiles data from many sources; cite the AI Index and its original source.
Next expected update
Yearly (Stanford HAI, yearly (April))
Original page
hai.stanford.edu: AI Index Report
Data files
CSV · JSON (stable links)
Full change log

No changes recorded yet. First checked at the source on 23 Oct 2026.

Cite this number

APA

Stanford HAI. (2026). 77.3% success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026). Via digitaltreed, https://digitaltreed.com/statistics/ai-agents/agent-success-terminal-bench

MLA

"77.3% success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026)." Stanford HAI, AI Index Report, 2026. digitaltreed, https://digitaltreed.com/statistics/ai-agents/agent-success-terminal-bench.

HTML link

<a href="https://digitaltreed.com/statistics/ai-agents/agent-success-terminal-bench">77.3% success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026)</a> (Stanford HAI, 2026)
Embed the chart
<iframe src="https://digitaltreed.com/statistics/embed/ai-agents-benchmarks" width="640" height="420" loading="lazy" title="AI agent success on benchmark tasks (Stanford AI Index 2026)"></iframe> <p>Source: <a href="https://digitaltreed.com/statistics/ai-agents/agent-success-terminal-bench">77.3% success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026), digitaltreed</a></p>
CSV

Keep the source link under the chart. The chart updates when the number does.

FAQ

Questions about this number

What is the latest figure for capability (Global)?

77.3% success rate for AI agents on real-world tasks in Terminal-Bench, up from 20% in 2025 (Stanford AI Index 2026), for 2026, according to Stanford HAI, AI Index Report. We checked it at the source on 23 Oct 2026.

Get the numbers when they change

One email a day at most, only when a source publishes new data. No other mail.

Put this number to work

We help small and mid-size firms turn data like this into useful content, answers for customers and visibility in AI search.

Talk to us

Every number on this page was checked at its original source. Free to cite with a link to this page.