Agentic & Tool Use benchmark

# OSWorld 2.0 leaderboard

> OSWorld 2.0 results for 9 AI models, led by Claude Opus 5 at 31.4%. What the benchmark measures, who runs it, and a source for every score.
- Canonical page: https://noometry.com/benchmarks/osworld-2
- Last updated: 2026-10-10
- Title: OSWorld 2.0 Leaderboard (October 2026): Scores by Model

As of October 2026, Claude Opus 5 has the highest published OSWorld 2.0 score on Noometry at 31.4%, out of 9 models with results.

Last verified October 10, 2026

## About OSWorld 2.0

The second version of the OSWorld computer-use benchmark.

- **Category:** [Agentic & Tool Use](https://noometry.com/best/agentic)
- **Introduced:** 2026
- **Format:** Computer use
- **Unit:** Percent (random guessing ≈ 0%)
- **Official site:** [os-world.github.io](https://os-world.github.io)

## Top 9 models

Top models on OSWorld 2.0

1.  Claude Opus 5 31.4%
2.  GPT-5.6 Sol 27.3%
3.  Claude Opus 4.8 20.6%
4.  Claude Opus 4.7 18.2%
5.  GPT-5.5 13%
6.  Claude Sonnet 4.6 9.3%
7.  Kimi K2.6 4.6%
8.  MiniMax-M3 4.6%
9.  Qwen3.7 Plus 2.8%
10.  010203040

Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise)

## All results

OSWorld 2.0 results by model
| # | Model | Provider | Score | Setting | Source | Date |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [Claude Opus 5](https://noometry.com/models/claude-opus-5) | [Anthropic](https://noometry.com/providers/anthropic) | 31.4% | max | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 2 | [GPT-5.6 Sol](https://noometry.com/models/gpt-5-6-sol) | [OpenAI](https://noometry.com/providers/openai) | 27.3% | max | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 3 | [Claude Opus 4.8](https://noometry.com/models/claude-opus-4-8) | [Anthropic](https://noometry.com/providers/anthropic) | 20.6% | max | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 4 | [Claude Opus 4.7](https://noometry.com/models/claude-opus-4-7) | [Anthropic](https://noometry.com/providers/anthropic) | 18.2% | max | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 5 | [GPT-5.5](https://noometry.com/models/gpt-5-5) | [OpenAI](https://noometry.com/providers/openai) | 13% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 6 | [Claude Sonnet 4.6](https://noometry.com/models/claude-sonnet-4-6) | [Anthropic](https://noometry.com/providers/anthropic) | 9.3% | medium | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 7 | [Kimi K2.6](https://noometry.com/models/kimi-k2-6) | [Moonshot AI](https://noometry.com/providers/moonshot) | 4.6% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 8 | [MiniMax-M3](https://noometry.com/models/minimax-m3) |  [![](/logos/minimax.svg) MiniMax](https://noometry.com/providers/minimax) | 4.6% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |
| 9 | [Qwen3.7 Plus](https://noometry.com/models/qwen3-7-plus) |  [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 2.8% |  | [Epoch AI](https://epoch.ai/benchmarks) |  |

## Compare the leaders

-   [Claude Opus 5 vs GPT-5.6 Sol](https://noometry.com/compare/claude-opus-5-vs-gpt-5-6-sol)
-   [Claude Opus 5 vs Claude Opus 4.8](https://noometry.com/compare/claude-opus-4-8-vs-claude-opus-5)
-   [Claude Opus 5 vs Claude Opus 4.7](https://noometry.com/compare/claude-opus-4-7-vs-claude-opus-5)
-   [Claude Opus 5 vs GPT-5.5](https://noometry.com/compare/claude-opus-5-vs-gpt-5-5)
-   [GPT-5.6 Sol vs Claude Opus 4.8](https://noometry.com/compare/claude-opus-4-8-vs-gpt-5-6-sol)
-   [GPT-5.6 Sol vs Claude Opus 4.7](https://noometry.com/compare/claude-opus-4-7-vs-gpt-5-6-sol)

## Other agentic & tool use benchmarks

-   [Terminal-Bench](https://noometry.com/benchmarks/terminal-bench)
-   [APEX-Agents](https://noometry.com/benchmarks/apex-agents)
-   [Berkeley Function Calling Leaderboard](https://noometry.com/benchmarks/bfcl)
-   [GDPval](https://noometry.com/benchmarks/gdpval)
-   [Remote Labor Index](https://noometry.com/benchmarks/remote-labor-index)
-   [TheAgentCompany](https://noometry.com/benchmarks/the-agent-company)
-   [τ²-bench Airline](https://noometry.com/benchmarks/tau2-airline)
-   [τ²-bench Banking](https://noometry.com/benchmarks/tau2-banking)
-   [τ²-bench Retail](https://noometry.com/benchmarks/tau2-retail)
-   [τ²-bench Telecom](https://noometry.com/benchmarks/tau2-telecom)
-   [Cybench](https://noometry.com/benchmarks/cybench)
-   [DeepResearch Bench](https://noometry.com/benchmarks/deepresearch-bench)

## Frequently asked questions

### What does OSWorld 2.0 measure?

The second version of the OSWorld computer-use benchmark.

### Which model has the highest OSWorld 2.0 score?

As of October 2026, Claude Opus 5 has the highest published OSWorld 2.0 score on Noometry at 31.4%, out of 9 models with results.

### What is the best open-weight model on OSWorld 2.0?

Kimi K2.6 has the highest OSWorld 2.0 accuracy among open-weight models at 4.6%, ranking 7 of 9 overall.

### Cite this page

Noometry. (2026). OSWorld 2.0 leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/osworld-2

Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/osworld-2.md).