Model card
Claude Computer Use.
AnthropicproprietaryUnknown paramsClaude 3.5 Sonnet with computer use tool
Claude 3.5 Sonnet (computer use beta, Oct 2024). Operates screenshot-based GUI. anthropic.com/computer-use.
§ 02 · Benchmarks
Every benchmark Claude Computer Use has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | OSWorld | Agentic AI · Web & Desktop Agents | success-rate | 22.0% | #21 | — | source ↗ |
| 02 | OSWorld | Agentic AI · Web & Desktop Agents | success-rate | 14.9% | #24 | 2024-04-11 | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where Claude Computer Use actually performs.
§ 04 · Papers
1 paper with results for Claude Computer Use.
- 2024-04-11· Agentic AI· 1 result
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
§ 05 · Related models
Other Anthropic models scored on Codesota.
§ 06 · Sources & freshness
Where these numbers come from.
anthropic-blog
1
result
arxiv
1
result
2 of 2 rows marked verified.