AI IQ
Subscribe

Computer Use

A bird's-eye view of the dimension: the main IQ chart, the effective-cost frontier, and the benchmark charts that feed it.

Computer Use IQ
Computer Use IQ
Each model's Computer Use IQ plotted on a standard normal IQ distribution
Controls:

How to read this chart

Models are placed on a standard IQ-style distribution using Computer Use IQ. Higher scores appear farther to the right.

Benchmarks in this dimension

Terminal-Bench HardTerminal-Bench 2.1BrowseCompMCP AtlasToolathlonOSWorld-VerifiedAgents' Last Exam
Computer Use vs Effective Cost
Computer Use vs Effective Cost
X = effective cost (log). Y = Computer Use IQ. Color = provider.
Controls:
Computer Use1:1Cost

How to read this chart

Each point is a public model. The chart compares Computer Use IQ against Effective Cost (per 1M I/O Tokens), with color showing the model provider.

Data sources
Artificial AnalysisARC PrizeVals.aiArtificial Analysis Terminal-Bench v2.1Vals.ai Terminal-Bench 2.1LLM StatsOSWorldanthropic-system-cardToolathlonScale LabsAgents' Last Exam
Benchmarks In This Dimension
Terminal-Bench Hard Benchmark Scores
Each model's Terminal-Bench Hard score. Color = provider.
Controls:

How to read this chart

Bars rank models by the source-backed benchmark value used for this chart. Longer bars indicate higher published scores.

Data sources
Artificial Analysis
Terminal-Bench 2.1 Benchmark Scores
Each model's Terminal-Bench 2.1 pass@1 score, using Artificial Analysis as canonical and Vals.ai as fallback. Color = provider.
Controls:

How to read this chart

Bars rank models by the source-backed benchmark value used for this chart. Longer bars indicate higher published scores.

Data sources
Artificial Analysis Terminal-Bench v2.1Vals.ai Terminal-Bench 2.1
BrowseComp Benchmark Scores
Each model's BrowseComp score. Color = provider.
Controls:

How to read this chart

Bars rank models by the source-backed benchmark value used for this chart. Longer bars indicate higher published scores.

Data sources
LLM Stats
MCP Atlas Benchmark Scores
Real-world MCP tool-use pass rates. Color = provider.
Controls:

How to read this chart

Bars rank models by the source-backed benchmark value used for this chart. Longer bars indicate higher published scores.

Data sources
Scale Labsanthropic-system-card
Toolathlon Benchmark Scores
Each model's Toolathlon score. Color = provider.
Controls:

How to read this chart

Bars rank models by the source-backed benchmark value used for this chart. Longer bars indicate higher published scores.

Data sources
LLM StatsToolathlonanthropic-system-card
OSWorld-Verified Benchmark Scores
Each model's OSWorld-Verified score. Color = provider.
Controls:

How to read this chart

Bars rank models by the source-backed benchmark value used for this chart. Longer bars indicate higher published scores.

Data sources
LLM StatsOSWorldanthropic-system-card
Agents' Last Exam Benchmark Scores
Full / Overall pass rates from Agents' Last Exam. Color = provider.
Controls:

How to read this chart

Bars rank models by the source-backed benchmark value used for this chart. Longer bars indicate higher published scores.

Data sources
Agents' Last Exam