activeAgents

OSWorld

Metric: Score (higher is better)

Real-world computer use tasks in desktop OS environments

Models ranked

5

tracked on this benchmark

Score band

79.6 – 62.6

best → lowest tracked

Snapshot trend

+5.00

Oct 17 → Oct 26 · 1 models

Leaderboard

Tracked models ranked by Score (higher is better).

Compare candidates
#Model variant and provenanceScore
1
Version: OSWorld V2.1; Kimi Agent harness; max reasoning effortHarness: Not recordedEvaluator: Not recordedObserved: Oct 23, 2026Confidence: Not recordedSource

Notes: Moonshot self-reported launch table, row 'OSWorld V2.1', column 'Kimi K3.1 Pro (max)'. Settings: reasoning effort max, temperature 1.0, top-p 0.95 (blog footnote). Footnote: 'evaluated with the Kimi Agent harness in the official OSWorld v2 repo'.

79.6
2

Configuration: gpt-6.1-terra (Max)

Version: OSWorld 2.1, offline set; max reasoning effortHarness: Not stated in the launch postEvaluator: OpenAIObserved: Oct 26, 2026Confidence: self-reportedSource

Notes: OpenAI launch post chart 'OSWorld 2.1, offline set' (Max 0.742; Low 0.538, Medium 0.667, High 0.687, Xhigh 0.713). Offline set, harness and run count not stated. Retrieved 2026-10-26.

74.2
3

Configuration: Gemini 4 Argon; max of 3 runs

Version: OSWorld 2.0 offline subset (partial score)Harness: Gemini CUA harness; official OSWorld 2.0 evaluatorEvaluator: Google DeepMindObserved: Oct 17, 2026Confidence: self-reportedSource

Notes: Blog table + methodology page: OSWorld 2.0 offline subset, partial score, max over 3 runs, Gemini CUA harness, official evaluator, 1080p, 500 steps, 08.08 patch. Comparability caveat: max-of-3 and offline subset differ from other OSWorld 2.0 rows. Retrieved 2026-10-17.

69.2
4
Version: OSWorld-VerifiedHarness: Not recordedEvaluator: Not recordedObserved: Feb 5, 2026Confidence: Not recordedSource
64.7
5
Version: OSWorld 2.0; computer useHarness: Not recordedEvaluator: Not recordedObserved: Jul 9, 2026Confidence: Not recordedSource

Notes: Official GPT-5.6 GA launch benchmark row.

62.6

How to read this benchmark

This benchmark scores models where higher is better. Use scores for directional filtering and shortlisting, not universal quality ranking; then validate pricing, context window, provider availability, and fit for your workload.

Trust this score when

  • There is a fresh timestamped snapshot (or multiple snapshots) for this benchmark.
  • The model list covers the same version family you can actually deploy today.
  • Top candidates overlap with your required routing and feature requirements.

Be cautious when

  • There is only one benchmark snapshot or the dataset appears stale.
  • The benchmark metric direction is opposite of your decision objective.
  • The score difference between options is narrow and likely within implementation variance.

Related benchmarks

Last reviewed: May 20, 2026

Resources